i Ta tN Tia aE
hi Me act ea AT
I agg he ole i
Ranta es <a e
ah es
Gah An Mra. pone eahn thn act
I ea ee ee
AS a Cc,” gt Wg Aig UR Oh a we > Ne Wy Ea
try JOO oy Ne Or gO Oy” ra SiC gO Oe
wnt” NY Nn. Sea” tp, SES
it]
Ma
Sr ‘y ord ay a aa &
LA botc Ray Lie os,
| i i i 7%
il | Hy ail “ee
- | | a a
| - | |
i er in I
Bee Ne Bp cPRsOT oe 4
; An Ne aS
Si, ea)
i
% a
Aes 4 oO
fi ats ‘Ss
\ es :
SWiceh ar Tol
Nee bee
ie? Rate
™ cathe"
& ay ve SS
-
on
raat
Sy
6)
eA
heair
ey (omy
ae chy
gp OR So
at e'
é xe { a)
— 4 ih oy AU =)
O < 4 BI ve “ty, li Wht
ees saul | I x “eg l
>
fot in diet oN
fe eG a aa
“Hoy te Nase 08
aN I Si Ces, | 1 A)
NT ad oe a va oa eg e : i AG t
EL atria. | Sahar ney e
Cae ney ee, Ny ey
aoe m, SOTO on < e dalle eae fa}
|
nae
Peto oo re
erat Bae”
taaenrt
ny
3
ah
A.
a
Fo
Pr oa
ar
car
Gham ts
te
a
ae:
Pay Di
ra
fe
toe
fa
jt
a =
iL
we
oa
iS i J
A Kogee tiakiD.
ET Soe ee Peed ae
i ne,
Ney
ee R
hea
NEF Mee gS
eeineae
SSL a
ane ee
3 tt
anny
&
os
x |
|
SB AT ee
Siero
mins 8 Ray
(ess ms
PS
‘Ak
Sretss7
Ae
i
sae]
=
tt
fe
ee | a
bat I 55
sta:
A
i
JOURNAL OF GEOLOGY
THE UNIVERSITY OF CHICAGO PRESS
CHICAGO, ILLINOIS
THE CAMBRIDGE UNIVERSITY PRESS
LONDON AND EDINBURGH
THE MARUZEN-KABUSHIKI-KAISHA
TOKYO, OSAKA, KYOTO, FUKUOKA, SENDAI
THE MISSION BOOK COMPANY
SHANGHAI
TEE
MOURNAL OF GEOLOGY
A Semi-Quarterly Magazine of Geology and
elated «Sciences
EDITED BY
THOMAS C. CHAMBERLIN AND ROLLIN D. SALISBURY
With the Active Collaboration of
SAMUEL W. WILLISTON ALBERT JOHANNSEN
Vertebrate Paleontology Petrology
STUART WELLER ROLLIN T. CHAMBERLIN
- Invertebrate Paleontology Dynamic Geology
ALBERT D. BROKAW, Economic Geology
ASSOCIATE EDITORS
SIR ARCHIBALD GEIKIE, Great Britain HENRY S. WILLIAMS, Cornell University
CHARLES BARROIS, France JOSEPH P. IDDINGS, Washington, D.C.
ALBRECHT PENCK, Germany JOHN C. BRANNER, Leland Stanford University
HANS REUSCH, Norway RICHARD A. F. PENROSE, Jr., Philadelphia, Pa.
GERARD DEGEER, Sweden WILLIAM H. HOBBS, University of Michigan
T. W. EDGEWORTH DAVID, Australia FRANK D. ADAMS, McGill University
BAILEY WILLIS, Leland Stanford Junior CHARLES K. LEITH, University of Wisconsin
University WALLACE W. ATWOOD, Harvard University
GROVE K. GILBERT, Washington, D.C. WILLIAM H. EMMONS, University of Minnesota
CHARLES D. WALCOTT, Smithsonian ARTHUR L. DAY, Carnegie Institution
Institution
VOLUME XXV
JANUARY-DECEMBER, 1917
Zs <ahsonian sti
p
THE UNIVERSITY OF CHICAGO PRESS
CHICAGO, ILLINOIS
Published
February, March, May, June, August, September,
November, December, 1917.
Composed and Printed By
The University of Chicago Press
Chicago, Illinois, U.S.A.
CONTENTS OF VOLUME X XV
NUMBER I
SYMPOSIUM ON THE AGE AND RELATIONS OF THE Fosstt HUMAN
REMAINS FOUND AT VERO, FLORIDA :
PAGE
EDITOR TAP ONODE iiss yuaede sac agaernel en i es Me aN Mie aiaell ie Ys I
ON THE ASSOCIATION OF HUMAN REMAINS AND EXTINCT VERTE-
BRATES AT VERO, FLORIDA. E. H. Sellards Bier tiger hae 4
INTERPRETATION OF THE FORMATIONS CONTAINING HUMAN BONES
AT VERO, FLORIDA. Rollin T. Chamberlin Sy \ es aa 25
ON REPORTED PLEISTOCENE HUMAN REMAINS AT VERO, FLORIDA.
snhomass Wayland Vaughan tne ee feiss goa Aa 40
PRELIMINARY REPORT ON FINDS OF SUPPOSEDLY ANCIENT HUMAN
REMAINS AT VERO, FLORIDA. Ales Hrdlicka j : i ‘ 43
THE QUATERNARY DEPOSITS AT VERO, FLORIDA, AND THE VERTE-
BRATE REMAINS CONTAINED THEREIN. Oliver P. Hay... 52
ARCHAEOLOGICAL EVIDENCES OF MAN’S ANTIQUITY AT VERO,
FLORIDA. Georre GrantMiacCurdy 1 ey ea) he ce ae 56
SUGGESTIONS FOR A QUANTITATIVE MINERALOGICAL CLASSIFICATION OF
Icneous Rocks. Albert Johannsen . 3 : A : ; 63
REVIEWS . ; : : : : i 4 i ; ; é 5 : 98
NUMBER II
ON THE HypotuHEsis oF Isostasy. W.D. MacMillan . . . .~ 105
THE Mippite PALreozoic STRATIGRAPHY OF THE CENTRAL ROCKY
MOUNTAIN REGION: 1. C2 We Tomlinson 77 oo ei LE
SoME Factors AFFECTING THE DEVELOPMENT OF Mup-Cracks. E. M.
Kindle ‘ : : i f 2 ; : : ; : : AP AREAS
DOWNWARPING ALONG JOINT PLANES AT THE CLOSE OF THE NIAGARAN
AND ACADIAN. Lancaster D. Burling . i SUS asia, MRR gad i) AGA cu ARG
THE WESTERN INTERIOR GEOSYNCLINE AND ITs BEARING ON THE ORIGIN
AND DISTRIBUTION OF THE COAL MEASURES. Francis M. Van Tuyl 150
A DECIMAL GROUPING OF THE PLAGIOCLASES. F. C. Calkins , : 157
STUDIES FOR STUDENTS: A CLASSIFICATION OF BrecciAs. W. H.
Norton vs if p i f A f : y ‘ : 5 160
REVIEWS . k : aN i : 4 % f : ; : : 195
vi CONTENTS OF VOLUME XXV
NUMBER III
THE PROBLEM OF THE ANORTHOSITES. N. L. Bowen
THE MippLE PALEoOzoIc STRATIGRAPHY OF THE CENTRAL ROCKY
MountTAIn ReEcion. II. C. W. Tomlinson j
A Few INTERESTING PHENOMENA ON THE ERUPTION OF USU.
Y. Oinouye Re armpit eyC a Hi
INTRAFORMATIONAL PEBBLES IN THE RICHMOND GROUP, AT WINCHES-
TER, Onto. August F. Foerste
REVIEWS
NUMBER IV
LABIDOSAURUS CoPpE, A LOWER PERMIAN COTYLOSAUR REPTILE FROM
Texas. Samuel W. Williston
NOTES ON THE 1916 ERUPTION OF Mauna Loa. Iand II. Harry O.
Wood
AGE AND STRATIGRAPHIC RELATIONS OF THE OLENTANGY SHALE OF
CENTRAL OHIO, WITH REMARKS ON THE PROUT LIMESTONE AND
SO-CALLED OLENTANGY SHALES OF NORTHERN OHIO. Amadeus
W. Grabau
Tue History oF Devit’s LAKE, Wisconsin. Arthur C. Trowbridge .
THe MippLE PALEOzoIC STRATIGRAPHY OF THE CENTRAL ROCKY
Mountain Recion. III. C. W. Tomlinson
REVIEWS
NUMBER V
Tue Laws or Exvastico-Viscous Frow. A. A. Michelson :
THE PHYLOGENY AND CLASSIFICATION OF REPTILES. S. W. Williston .
Our PRESENT KNOWLEDGE OF ISOSTASY FROM GEODETIC EVIDENCE.
William Bowie .
THE Satsop FORMATION OF OREGON AND WASHINGTON. J Harlen
Bretzi-
THE CORROSIVE ACTION OF CERTAIN BRINES IN MANITOBA. R. C.
Wallace
NOTES ON THE 1916 ERUPTION OF Mauna Loa. IIIT andIV. Harry O.
Wood
A PRoposepD Diep Protractor. Chester K. Wentworth
PETROLOGICAL ABSTRACTS AND Reviews. Albert Johannsen —
RECENT PUBLICATIONS
PAGE
200
244
258
289
397
309
322
337
344
373
395
405
4It
422
446
459
467
489
492
498
CONTENTS OF VOLUME XXV
NUMBER VI
STRUCTURE OF THE ANORTHOSITE BODY IN THE ADIRONDACKS. H. P.
Cushing .
ADIRONDACK INTRUSIVES. N. L. Bowen
ADIRONDACK INTRUSIVES. H. P. Cushing
A REVIEW OF THE AMORPHOUS MINERALS. Austin F. Rogers
THE CHAMPLAIN SEA IN THE LAKE Ontario Basin. Kirtley F.
| Mather
THE RELATIONSHIPS OF THE FossiL Brrp Palacochendides Mioceanus.
Alexander Wetmore
A STUDY OF THE FAUNAS OF THE RESIDUAL MISSISSIPPIAN OF PHELPS
County (CENTRAL OzARK REGION), Missourr. Josiah Bridge
EVIDENCE BEARING ON A POSSIBLE NORTHEASTWARD EXTENSION OF
MISSISSIPPIAN SEAS IN ILLINoIs. W. W. Davis
DISCUSSION OF “‘SOME EFFECTS OF CAPILLARITY ON Ort ACCUMULA-
TION,” By A. W. McCoy. C. W. Washburne
PETROLOGICAL ABSTRACTS AND REviEws. Albert Johannsen
REVIEWS
NUMBER VII
ON THE AMOUNT OF INTERNAL FRICTION DEVELOPED IN ROocKS
DURING DEFORMATION AND ON THE RELATIVE PLASTICITY OF
DIFFERENT Types OF Rocks. Frank D. Adams and J. Austen
Bancroft
ON THE MATHEMATICAL THEORY OF THE INTERNAL FRICTION AND
LIMITING STRENGTH OF ROCKS UNDER CONDITIONS OF STRESS
EXISTING IN THE INTERIOR OF THE EartH. Louis Vessot King
NOTE ON THE DEPOSITS CONTAINING HUMAN REMAINS AND ARTIFACTS
AT VERO; Ftoripa. E. H. Sellards
Tue Fossit PLANTS FROM VERO, FLtormwA. Edward W. Berry
FURTHER STUDIES AT VERO, FLormpa. Rollin T. Chamberlin
ANOTHER LOCALITY OF EOCENE GLACIATION IN SOUTHERN COLORADO.
Wallace W. Atwood
REVIEWS
RECENT, PUBLICATIONS
594
Vili CONTENTS OF VOLUME XXV
NUMBER VIII
Tue ACTIVE VOLCANOES OF NEW ZEALAND. E.'S. Moore
FOOTHILLS STRUCTURE IN NORTHERN CoLorapbo. Victor Ziegler .
ON THE GEOLOGY OF THE ALKALI ROCKS IN THE TRANSVAAL.
Brouwer : : : : : ‘ ! : i
PETROGRAPHICAL ABSTRACTS AND REviEws. Albert Johannsen
REVIEWS
RECENT PUBLICATIONS
INDEX TO VOLUME XXV
H. A.
PAGE
693
715
741
779
782
788
789
VOLUME XXV NUMBER tr
THE
JOURNAL or GEOLOGY,.
A SEMI-QUARTERLY ZL ent a Ne
_ EDITED By (2 FOt2 2
THOMAS C. CHAMBERLIN AND ROLLIN D. SALISBURY X Noses Sh eg
With the Active Collaboration of = a a
SAMUEL W. WILLISTON, Vertebrate Paleontology ALBERT JOHANNSEN, Petrology
STUART WELLER, Invertebrate Paleontology ROLLIN T. CHAMBERLIN, Dynamic Geology"
ALBERT D, BROKAW, Economic Geology
ASSOCIATE EDITORS Re We
SIR ARCHIBALD GEIKIE, Great Britain JOSEPH P.IDDINGS, Washington, D.C. Sa
CHARLES BARROIS, France JOHN C, BRANNER, Leland Stanford Junior University
ALBRECHT PENCK, Germany RICHARD A. F. PENROSE, Jr., Philadelphia, Pa.
_ HANS REUSCH, Norway WILLIAM B. CLARK, Johns Hopkins University
_ GERARD DEGEER, Sweden . WILLIAM H. HOBBS, University of Michigan
* Tf. W. EDGEWORTH DAVID, Australia FRANK D. ADAMS, McGill University
_ BAILEY WILLIS, Leland Stanford Junior University CHARLES K. LEITH, University of Wisconsin
GROVE K. GILBERT, Washington, D.C. WALLACE W. ATWOOD, Harvard University
CHARLES D. WALCOTT, Smithsonian Institution WILLIAM H. EMMONS, University of Minnesota
HENRY S. WILLIAMS, Cornell University ARTHUR L. DAY, Carnegie Institution
Bord . ‘:
°
JANUARY-FEBRUARY 1917
SYMPOSIUM ON THE AGE AND RELATIONS OF THE FOSSIL HUMAN REMAINS
FOUND AT VERO, FLORIDA:
- EpirortAL NOTE “= = = = soa > Sama ipl = - - - = Sins I
ON THE ARgOCIATLON oF HumAN REMAINS AND EXTINCT VERTEBRATES AT VERO, FLORIDA
‘ E. H. SELLARDS 4
oe ene eAmON OF THE FORMATIONS Contarninc Human Bones AT VERO, FLORIDA
Roun T. CHAMBERLIN 25
ON REPORTED PLEISTOCENE HumAN REMAINS AT VERO, FLORIDA —
THOMAS WAYLAND VAUGHAN 40
Saran e ReEporT ON FINDS OF SUPPOSEDLY ANCIENT HumMaN REMAINS AT VERO,
FLORIDA —- = Aha - - = - - - - E ALES HRDLICKA 43
| THe QuaTERNARY Deposits AT VERO, FLORIDA, AND THE VERTEBRATE REMAINS CON-
| TAINED THEREIN - = -— - - - - = ae - - OLIVER P. Hay 52
2 ARCHAEOLOGICAL EVIDENCES OF MAN’s ANTIQUITY AT VERO, FLORIDA
GrorcE Grant MacCurpy 50
SUGGESTIONS FOR A QUANTITATIVE MINERALOGICAL CLASSIFICATION OF
IGNEOUS ROCKS - - - - - - - - ALBERT JOHANNSEN 63
REVIEWS = AiG eae ea cee nial poids ee - SNS aE SAA t ee TRE OES OE aint Malas)
Side UNIVERSITY OF CHIREAGO PRESS
CHICAGO, ILLINOIS, U.S.A.
AGENTS
Rt THE CAMBRIDGE UNIVERSITY PRESS, Lonpon anp EpInBuRGH
KARL W. HIERSEMANN, Letrzic :
THE MARUZEN-KABUSHIKI-KAISHA, Toxyo, Osaka, Kyoro, Fukuoka, SENDAI
* THE MISSION BOOK COMPANY, SHancuat
THE JOURNAL “OR “CEOLOG.
EDITED BY
THOMAS C. CHAMBERLIN AND ROLLIN D. SALISBURY
With the Active Collaboration of
SAMUEL W. WILLISTON : ALBERT JOHANNSEN
Vertebrate Paleontology Petrology
STUART WELLER ROLLIN T. CHAMBERLIN
Invertebrate Paleontology . é: Dynamic Geology
ALBERT D. BROKAW
Economic Geology
The Journal of Geology is published by the University of Chicago at the University Press semi-
quarterly, on or about the following dates: February 1, March 15, May 1, June 15, August I, September 15,
November 1, December 15. {| The subscription price is $4.00 per year; the price of single copies is
65 cents. Orders for service of less than a half-year will be charged at the single-copy rate. {| Postage is
prepaid by the publishers on all orders from the United States, Mexico, Cuba, Porto Rico, Panama Canal
Zone, Republic of Panama, Hawaiian Islands, Philippine Islands, Guam, Samoan Islands, Shanghai.
4] Postage is charged extra as follows: For Canada, 30 cents on annual subscriptions (total $4.30), on
single copies, 4 cents (total 60,cents); for all other countries in the Postal Union, 53 cents on annual
subscriptions (total $4.53), on single copies, 11 cents (total 76 cents). [Patrons are requested to make
all remittances payable to the University of Chicago Press in postal or express money orders or bank >
drafts.
The following agents have been appointed and are authorized to quote the prices indicated:
For the British Empire: The Cambridge University Press, Fetter Lane, London, E.C. Yearly
subscriptions, including postage, 18s. 8d. each; single copies, including postage, 35. 2d. each.
For the continent of Europe: Karl W. Hiersemann, Konigstrasse 29, Leipzig, Germany. Yearly
subscriptions, including postage, M. 22.65 each; single copies, including postage, M. 3.80 each.
For Japan and Korea: The Maruzen-Kabushiki-Kaisha, 11 to 16 Nihonbashi Tori Sanchome,
Tokyo, Japan. Yearly subscriptions, including postage, Yen 9.00 each; single copies; including
postage, Yen 1.50 each.
For China: The Mission Book Chaat 18 Peking Road, Shanghai. Yearly subscriptions,
#4.00; single copies, 65 cents, or their equivalents in Chinese money. Postage extra, if mailed direct
outside of Shanghai, on yearly subscriptions 53 cents, on single copies I1 cents.
Claims for missing numbers should be made within the month following the regular month of
publication. The publishers expect to supply missing numbers free only when they have been lost in
transit.
Business correspondence should be addressed to The University of Chicago Press, Chicago, Ill.
Communications for the editors and manuscripts should be addressed to the Editors of THE
JOURNAL OF GEOLOGY, the University of Chicago, Chicago, Ill.
Contributors are requested to write scientific and proper names with patel care and in citations
to follow the form shown in the pages of the Journal.
Separates, if desired, must be ordered in advance of publication. Separates of leading articles will be
printed as ordered, of which 50 (without covers) will be furnished gratis, the actual cost of the remainder
(and covers, if desired) to be paid for by the author. The actual cost may vary, and will depend upon
the amount of work in re-making the pages into forms, presswork, paper, binding, etc. Separates
containing half-tones may be expected to cost somewhat more, the increase Hones upon the number
of cuts and the amount of work required upon them.
4
Entered as second-class matter, March 20, 1893, at the Post-office at Chicago, Ill., under the Act of March 3, 1879.
SRE SES Bee iia ik les aaaae da
VOLUME XXV NUMBER I
THE
JOURNAL OF GEOLOGY
JANUARY-FEBRUARY 1917
SYMPOSIUM ON THE AGE AND RELATIONS OF
THE FOSSIL HUMAN REMAINS FOUND
AT VERO, FLORIDA
Dr. E. H. SELLARDs, State Geologist of Florida
Dr. RoLtIn T. CHAMBERLIN, Geologist, University of Chicago
Dr. T. WAYLAND VAUGHAN, Geologist in Charge of the Coastal Plain Investigations
of the United States Geological Survey
Dr. A. HrpiiéKa, Curator of Physical Anthropology, United States National Museum
Dr. O. P. Hay, Research Associate, Carnegie Institution of Washington
Dr. G. G. MacCurpy, Anthropologist, Yale University
Epiror1aL Note.—In the issue of the American Journal of
Science of July, 1916, Dr. E. H. Sellards, state geologist of Florida,
announced the discovery of fossil human bones and artifacts in
association with the relics of many extinct vertebrates in a stream
deposit near Vero on the east coast of Florida. In the issue of
Science of October 27, 1916, there appeared a supplementary
article, by the same author, giving additional data bearing on the
age and relations of these interesting remains. About the same
date there appeared a more comprehensive statement in the
Eighih Annual Report of the Florida Geological Survey. Soon after
the issuance of the first paper, Dr. Sellards submitted to the
editors of the Journal of Geology an additional article empha-
sizing certain aspects of the question of man’s relationship to the
I
2 SYMPOSIUM
extinct vertebrates not set forth with equal fulness in the previous
article. ‘The tender of this manuscript was accompanied by a
very cordial invitation to visit the deposits at Vero and make
independent examination. Similar invitations were extended to
representatives of the Smithsonian Institution, the National
Geological Survey, and other institutions and individuals interested
in the subject. This opportunity for co-operative inspection
before publication fell happily into the policy of the Journal,
especially as the crowded state of its columns did not permit
immediate publication. While the Journal of Geology does not
hold itself immediately responsible for the conclusions advanced
by its contributors, it desires, so far as possible, when the issues
are vital, that all tenable aspects of interpretation shall be placed
before its readers that they may form their own conclusions on the
amplest available basis.
A conference was finally arranged for the last of October, in
which there participated Dr. A. Hrdlicka, anthropologist of the
United States National Museum; Dr. T. Wayland Vaughan,
geologist in charge of the coastal plain investigations of the United
States Geological Survey; Dr. O. P. Hay, special student of
Pleistocene vertebrates; Dr. G. G. MacCurdy, anthropologist
of Yale University; and Dr. R. T. Chamberlin, as representative
of the Journal of Geology. The members of the conference en-
joyed the guidance and assistance of Dr. Sellards; his assistant,
Mr. H. Gunter; and his local colleagues, Mr. Isaac M. Weills and
Mr. Frank Ayers, whose courtesies were unbounded. The visits of
these special students were only partially concurrent, that of Dr.
Chamberlin extending from October 23 to 28, that of Dr. Hrdlicka
from October 25 to 30, that of Dr. Hay from October 25 to 31, that
of Dr. MacCurdy from October 25 to 29, and that of Dr. Vaughan
from October 27 to 30; hence, while all met upon the ground, their
examinations were largely independent. The present assemblage
of the several statements of these visiting investigators into a
symposium, in connection with the paper of Dr. Sellards—revised
after the conference—was arranged without specific knowledge
of the conclusions of any of the visiting parties, except, of course,
those of the Journal’s own representative, and the independence
FOSSIL HUMAN REMAINS AT VERO, FLORIDA 3
of the reports has been preserved in passing the manuscripts through
the press. The statements are arranged in the order of their
receipt.
The dates of issuance of this and the preceding number of the
Journal have been advanced, so that this important assemblage
of data might be in the hands of those specially interested before
the holiday meetings of the scientific societies, while, at the same
time, delay in publishing other waiting articles might be avoided.—
EDITORS.
ON THE ASSOCIATION OF HUMAN REMAINS AND
EXTINCT VERTEBRATES AT VERO, FLORIDA
E. H. SELLARDS
State Geologist of Florida, Tallahassee, Florida
CONTENTS
INTRODUCTION
SKETCH Map OF THE LOCALITY
DESCRIPTION OF THE SECTION INCLUDING STRATA NOS. I, 2, AND 3
HuMAN REMAINS AND ARTIFACTS FROM STRATUM NO. 2
HuMAN REMAINS AND ARTIFACTS FROM STRATUM NO. 3
FOSSILS FROM STRATUM NO. 1
FOSSILS FROM STRATUM NO. 2
FossiLs FROM STRATUM NO. 3
INTERPRETATION OF THE SECTION
RELATION OF THE HUMAN REMAINS TO THE ASSOCIATED FOSSILS
The presence of vertebrate fossils in deposits exposed near Vero
in eastern Florida first became known in 1913. Fossil human
remains were not found at this locality, however, until October,
t915. Subsequently additional human skeletal material was
obtained in February, April, and June, 1916. The associated
fossils, which are numerous and varied, have been collected practi-
cally continuously since the locality became known, although the
largest collections are those made in February, 1915, and in Feb-
ruary, April, June, October, and November, 1916. During the
latter part of October the writer and his associates in Florida
enjoyed and profited by the presence at this locality of Drs. Hay,
MacCurdy, Hrdlicka, Vaughan, and Chamberlin, all of whom are
participants in the present discussion. The writer is also per-
sonally indebted to the several specialists in different branches of
paleontology who have identified and described fossils from this
locality, acknowledgment of which is made in the subsequent pages
of this paper.
The discovery of human remains in association with extinct
vertebrates at this locality was announced by the writer in the
4
FOSSIL HUMAN REMAINS AT VERO, FLORIDA 5
issue of the American Journal of Science of July, 1916. Later dis-
coveries were described in Science in the issue of October 27, 1916.
Subsequently, in the Eighth Annual Report of the Florida Geological
Survey, published in October, 1916, the human remains and the
associated fossils were more fully described. The present paper
includes supplementary observations made during October and
November, 1916.
SKETCH MAP OF THE LOCALITY ~
The fossils at this place were found in a stream bed and were
discovered as the result of the construction of a drainage canal.
As an aid in interpreting the section through the stream bed the
reader may refer to the sketch map of the locality and surroundings
shown in Fig. 1. The chief topographic features include a Pleis-
tocene beach and the drainage system of the stream in which the
fossils were found. On the east is the narrow body of ocean water
known as Indian River and the beach of the present shore line.
The ancient beach at this place is low, having an elevation of from
5 to 15 feet above the adjoining flat lands. Both to the north and
to the south, however, the ridge formed by the beach becomes more
pronounced. ‘This beach, in fact, is a part of the extensive Pleis-
tocene barrier beach which approximately parallels the present
shore line for 200 or 300 miles in eastern Florida, and is comparable
in origin to the modern or existing ocean beach which lies from one
to six miles farther east. The land both in front and back of the
beach is prevailingly flat and presents but little variation in level.
Such minor elevations as are found tend to assume the form of ridges
with a general north-south trend, separated by slight intervening
depressions which not infrequently are imperfectly drained. A
pronounced north-south ridge or beach is found about 1o miles
inland and is known locally as Ten Mile Ridge.
The drainage system of the stream in which the fossils were
found is very limited in extent and is controlled largely by the
Pleistocene beach. The valley of the main stream, which has a
width of from 350 to soo feet, extends from tidewater in the
Indian River into, but not across, this beach. Near the place
where the fossils were found the broad valley terminates abruptly
6 E. H. SELLARDS
and receives a tributary from the north and another from the
south, each of which, however, is of very limited extent. The
tributary from the south reaches as far as the railroad station at
Vero, a distance of about a half-mile, and one prong also finds its
way across the beach and extends as an indefinite drain into the
lowlands a distance of possibly a mile. The tributary from the
north likewise divides: the west prong, crossing the beach, heads
less than a mile to the northwest, while the east prong, which does
not cross the beach, continues to the north, paralleling the beach
Fic. 1.—Sketch-map showing the locality near Vero from which fossil human
remains have been obtained. Scale, 1 inch=4,000 feet. No. 1, pine lands; No. 2,
Pleistocene beach; No. 3, stream valley. The human remains were found in the
canal bank in this valley, west of the railroad and of the public-road crossing.
to Gifford Station, a distance of about 13 miles. The whole drain-
age system is thus very limited, involving only a few square miles,
and is in striking contrast to the broad valley which the stream has
developed in its lower course. Owing to the breadth of the valley,
it may possibly be inferred that at some former time the stream
had a larger drainage basin than at present. This, however, does
not seem to have been the case, since a pronounced cut or stream
channel across the beach, if made, would have persisted to the
present time.
FOSSIL HUMAN REMAINS AT VERO, FLORIDA 7
The native vegetation is distinctive on the beach, on the flat-
lands, and in the stream valley. The beach is characterized by
spruce pine, Pinus clausa, and by an undergrowth of shrubs in
which evergreens predominate. ‘The flatlands support a scattered
growth of long-leaf pine, Pinus palustris, the undergrowth being
chiefly saw palmetto. In the stream valley is found a dense
timber growth consisting chiefly of hardwood deciduous trees and
the cabbage palmetto. The outlines of the valley and of the beach
may be very definitely followed by the vegetation, which is con-
trolled in turn by the soil and by the drainage conditions.
The drainage canal, which starts at sea-level on the Indian
River, extends due west about one mile before entering the valley
of the stream. After following the stream a distance of about
1,000 feet, and having passed under both the railroad and the
public road, the canal leaves the valley near the union of the two
tributaries, cuts through the beach, and extends inland in a general
southwesterly direction about 12 miles. ‘The water level in the
canal at low-water stage at the locality where the fossils were
found is probably not more than 1 foot above sea, although upon
crossing the beach the water level is lifted by means of a spillway
to approximately 11 feet above mean sea-level. The land surface
for a distance of 12 or 15 miles inland probably nowhere exceeds
an elevation of 20 or 25 feet, except Ten Mile Ridge, which is
34 feet above mean sea-level.
DESCRIPTION OF THE SECTION THROUGH THE STREAM VALLEY
INCLUDING STRATA NUMBERED I, 2, AND 3
The section through the stream valley, as exposed in the canal
bank, includes three more or less well-marked divisions, which in
the present, as in the preceding papers, may be numbered 1, 2, and
3, No. 1 being at the base of the section and No. 3 at the top. In
No. 3 of the section are found human remains and artifacts,
vertebrate, land and fresh-water invertebrate, and plant fossils.
In No. 2 are found human remains, flint spalls, and probably also
bone implements, as well as vertebrate, land and fresh-water
invertebrate, and plant fossils. From the basal member of the
section, a marine deposit, no human remains have been obtained.
8 E. H. SELLARDS
To what extent the three divisions of this section represent dis-
tinct time intervals, and, on the other hand, to what extent they
may intergrade and thus express continuity of time, is discussed
subsequently. ‘These divisions are sufficiently well marked to be
recognized throughout the greater part or all of the section, and
serve as convenient markers in the exact placing of fossils.
The marine deposit, No. 1 of the section, is common to this
part of the Atlantic Coast of Florida, and is known to extend both
to the north and to the south, being a part of an extensive shallow-
water marine formation which borders the Atlantic Coast in
Florida for a distance of 200 or 300 miles. This stratum is. pre-
vailingly a shell marl, although it contains considerable sand, and
in places may consist wholly of sand of medium fine texture. A
large exposure, however, will scarcely fail to reveal the presence
of the marine shells.
Stratum No. 2, on the other hand, is probably local, represent-
ing fill in the stream valley, although its time equivalent, as indi-
cated by the fauna, is found at many localities throughout the
state. This deposit in the stream valley averages 3 or 4 feet in
thickness, and consists chiefly of rather coarse sand, which at the
top as a rule grades into fresh-water marl. Within the stratum,
filling holes or channels in the underlying deposit, are found local
accumulations of muck, including often wood, sticks, acorns,
snail shells, and vertebrate fossils. As a rule the sand near the
base of this stratum is light-colored and distinctly cross-bedded,
the heavy minerals, including staurolite, ilmenite, and quartz,
being deposited in bands and in pockets according to the size of the
grain and the specific gravity of the minerals. From 2 to 3 feet
above the base of the stratum the sand loses its cross-bedding and
becomes dark in color, owing to the inclusion of organic matter.
At the top, as has been stated, the sand passes into marl, contain-
ing an abundance of land and fresh-water invertebrates.
Stratum No. 3 consists chiefly of layers of muck and vegetable
material, alternating with layers of loose, nearly pure, light-colored
sand. This alternation of sand and muck is both abrupt and
frequent, the layers in places having only a thickness of from one-
half to 2 or 3 inches. At the top this stratum grades into a fresh-
FOSSIL HUMAN REMAINS AT VERO, FLORIDA 9
water marl which, in places, reaches a thickness of 18 inches. The
maximum thickness of this stratum is about 5 feet, although its
average thickness is from 2 to 3 feet.
The accompanying sketch, Fig 2, shows the section exposed
in the north bank of the canal from the railroad bridge west for
a distance of 500 feet. No. 1 is the marine shell marl; No. 2 is the
sand stratum which at the base is cross-bedded and at the top
passes into fresh-water marl; No. 3 is the deposit of muck and
vegetable material with alternating layers of incoherent sand.
The letter 6 indicates the location of one of the holes or channels
in the shell marl containing muck and driftwood as well as verte-
brate, invertebrate, and plant fossils.
In Fig. 3 is shown a section, drawn to scale, of 75 feet of the
south bank of the canal, showing the exposure as seen in November,
1916. ‘This section includes that part of the bank west of the
entrance of the lateral canal from the south, and thus passes
through the exposure at which some of the important fossils have
been found. Stratum No. 1 has an approximately even top sur-
face, although at one place near the middle of the section it is cut
into rather deeply by stratum No. 2. - This place, in fact, repre-
sents another of the holes or channels in No. 1 filled with muck
and decayed wood. Stratum No. 2 is variable in thickness, being
cut into at places by stratum No. 3. Stratum No. 3 as seen in this
section is variable both in thickness and in lithologic character-
istics. Its maximum thickness near the middle of the section is
about 5 feet, the upper 18 inches of which is a fresh-water marl.
The top or ground surface of this stratum is cut into at a and at b.
The cut at a@ was probably made in connection with dredging
operations. That at 6, however, is evidently the channel of the
modern stream where it cut into stratum No. 3.
At the point f in this section the muck and alluvial material of
No. 3 grades laterally by an indefinite line into the marl rock. In
the writer’s former papers the whole section at e was referred to
stratum No. 2, No. 3 being interpreted as absent at this place.
The present exposure apparently indicates that the two feet of
marl at e is the equivalent of the muck and marl bed of No. 3.
A similar section is seen on the opposite or east side of the lateral
‘PPUULYD WeIT}S UIIPOU 9Y} JO WOT}BIO] 9} S9BOIPUT Q 7e 4nd ay} ‘suoT}eIodo BuISperp oy} YIM UoTIIUTIOD
ur opeut ATqissod sem p ye ynd ay], *Y Ye Ua—s SI ‘E ‘ON WN}eI]S JO ssaUyOIY} [[N} OY sv [JOM se ‘Ios [PUISIIO oy, “[AVUT J9}eM-YSeIj OY} 0FUT
Aye19ze] sopess wnjzerys sty} f Aq pozeorpur sovyd oy} IeoN “pIeu JoyeM-Ysory ev oWUT do} oY} 1% Surpess pue ‘yonur puv purs UsrIayoour Jo
s19Av] SuIyeUIa}e Jo Suysisuoo ysodap payers ‘€ ‘oN ‘pues JozeM-Ysoay ‘c ‘ON {[reUr [[oys-outrew ‘I ‘ON ‘Jooy TI=YOUT I ‘ayeos Te}UOZ
“Hoy pur peonsap “qooy 44 Jo VouR}SIP B “4soM [UBD [eI9}¥] OY} Jo soURIZUA 94} Wo [eULd oY} Jo YUeG YINOs at UO UONIEG6—'E ‘ory
PRADA PPANARAARINNARARRRP AAR DaADT
VAI AAAIAAD DAD TFIIAAN FOIA AAIDNAAMN
PA AANADAAAQMP RAF ARR APRA ARR AR
PAA AMNIDAANDNNAN ANNAN ANANNINANR
E. H. SELLARDS
‘yooy Sc = YUL I “opeos [eot}IOA fyooy C4 = Your I ‘apeos [eJWOZTIOFY *¢ 7e PUNO] seM ss7am gnagns Jo uauitdeds adA} oy}
‘3 ye punoy sem ss4akv siupg Jo usumpeds ad} oy], “sfissoy [ewrue pue yuryd Surureyuoo yonut pue pues jUarayoour ‘f ‘oN ‘s[Issog oyeIq
“o}I9A pue Guefd ‘ayerqoJ9AUl JoyeM-YsorJ pue puUe] Surureyuos ‘poq jeu pu ‘yonur ‘pues soyeM-Ysory ‘Zz “ON ‘pues pue [IeU [[oYs-ouLIeUT
‘ION “Jo9J ZIS Jo ouvysip v ‘Ysom aSpliq peoseY ysvoD sey epuOpy] oy} wory [eu oy} Jo YUeq YIU ay} UO worjpeaS—z “OL
= ES Te) PARE ys NN Naren ee = NNER Et on UR ry ale ra) Ea See Mes Ne a Wi a POE I
eaipSeantien amc ene a an ere ee ee as ATA SY Ea iota ee ee on ae Aa eae Pe eae pial Ca a ak eas See CoS SN Nts cata tebe Eee
EN en ee ee ee
Pp Ee Se ee nearer Sigel et Ee 6 PPO ALDOR De en i eS oe aA
recat sig sap fay Pesca coy ise @eteh ose BG ae Bea see ag tent cer aetescionetsenaesccerngte Qiencar are ee arin
PLT Se CPLA ECR ERE RR TOE ORE I on SIL ER eT EE nS POE Bran eane ta CL CMe ay faa
€ 09 0.05 0000 89088 69 0,6 60-8107 80988400 O San ac OO oh go CE re TE 20 Me we %, 08 Sp Of e00 28 300 eon Oa"e 00,0 0,00 900 020.0. aoen
° °
00,0 Peo Oooo 6 95° an O00 LS a obee— 0D So enon ° P.o°e of, ° ot 2% 4
Oo epee 0°05 09 2. U > 'o © POLSON OS tO Se §o°220'o
oon ee i 0g 0 905g 00 0 0 00.00%, om pe of =
IO
FOSSIL HUMAN REMAINS AT VERO, FLORIDA ua
canal, and apparently the marl rock of that section may also be
referred to stratum No. 3. This part of the section will be more
fully discussed subsequently.
The sketch, Fig. 2 of this paper, may be compared with Fig. 6
of the writer’s paper in the Highth Annual Report of the Florida
Geological Survey, in which is shown a part of the same bank,
including the location of the important fossils. Although this part
of the canal bank was afterward carried back by excavations a
distance of from 5 to 8 feet, the fossils of the former sketch may, as
a matter of convenience, be projected on to the present sketch,
as has been done in this figure, thus indicating their approximate
location with respect to the section as now exposed. Human
bones were found in No. 2 at c (projected from the former section)
and in No. 3 at the general locality indicated by d. Pottery and
bone implements are found in No. 3 throughout the section.
HUMAN REMAINS AND ARTIFACTS FROM STRATUM NO. 2
The first human bones obtained at Vero were found in the
south bank of the canal, 330 feet west of the bridge. In the
exposure at this place there is no recognizable break in the section
from the base of stratum No. 2 to the marl rock at the top of the
section, and in the writer’s earlier papers the whole section was
referred to stratum No. 2. The new observations recorded in
this paper apparently permit the reference of the marl rock
at the top of the section to stratum No. 3. If this is true, the
human bones at this place in stratum No. 2 are beneath the one
and one-half or two feet of marl rock which represents stratum
No. 32.
The second lot of human bones from stratum No. 2 were found
by the writer in June, 1916. The bones found in place include
an astragalus, a cuneiform, and a part of an ilium. Upon sifting
the sand in which these bones were imbedded there was obtained
in addition two phalanges, a section from a limb bone, and some
other human bone fragments. The cuneiform was about ro inches
from the astragalus, and between the two bones at the same level
as the astragalus was the scapula of a deer. The ilium was about
one foot farther back in the bank. The vertebrate fossils found
I2 VE ASE LEARDS.
in the bank at this locality have been listed in the papers pre-
viously published.
The flints obtained from stratum No. 2 include a spall found
in place 3 feet east of the human bones listed in the preceding
paragraph, and about one foot farther back in the bank. Upon
passing the sand through a sieve five additional flint spalls were
obtained from this stratum, one of which was found about 10 feet
SCALE 1 INCH = 4 FEET
Fic. 4.—Ground plan showing the location of human bones found in the canal
bank at Vero in April and in June, 1916. The canal bank at this place faces approxi-
mately northwest and those bones, the location of which is indicated by Nos. 1-5,
and also several of the skull fragments, were collected in April. Those bones num-
bered 6-9 and a few skull fragments were collected in June. Among bones not specially
numbered are a part of the right femur and an incisor tooth found near No. 9, and
a part of a metatarsal found near No. 2. The bones indicated by Nos. 1-5 and No. 9
and several of the skull fragments were found on or near the contact line of strata
Nos. 2 and 3. Those bones indicated by Nos. 6-8 were found in stratum No. 2.
Bones taken from the caving of the bank and from siftings are not shown in this plan.
Index to bones: 1, left ulna; 2, a part of the shaft of the same bone; 3, left femur;
9, a part of the shaft of the same bone; 4, radius; 5, metatarsal; 6, astragalus; 7, ex-
ternal cuneiform; 8, part of ilium. ‘The place in the bank of bones Nos. 1, 3, and 4,
and bones Nos. 6 and 7, is shown in photographs previously published.
farther west. From these siftings were obtained also one nearly
complete small bone implement and a piece of a second bone
implement. Markings on bones from this stratum, which may have
been made by tools, have been described in the papers to which
reference has been made.
FOSSIL HUMAN REMAINS AT VERO, FLORIDA 13
The spalls found in this stratum have been placed in the hands
of Dr. George Grant MacCurdy, who has consented to report upon
them. It is evident that these flints are not of local origin, the
nearest known outcrop of a flint-bearing formation being more
than too miles to the northwest. The flint spalls are sharp-
edged and quite unworn, indicating that they have been trans-
ported no great distance by water. ‘The small flint pebbles
occasionally found in this deposit, on the contrary, are rounded
and well worn, there being no intermediate stages between the
worn pebbles of the deposit and the sharp-edged spalls.
HUMAN REMAINS AND ARTIFACTS IN STRATUM NO. 3
Three finds of human skeletal material are reported from
stratum No. 3. In each instance the bones lay at the contact line
between strata Nos. 2 and 3, and hence may belong to No. 2 rather
than to No. 3. The first of these discoveries was in the south
bank of the canal at the locality shown in Fig. 3. The bones
obtained at this place include the right and left ulna, part of a
humerus, a scapula, two incisors, parts of right and left femora,
a radius, part of a jaw, two metatarsals, and a considerable portion
of the skull, the pieces of which were dissociated and scattered.
All of these bones were on or very near the contact line between
Nos. 2 and 3, and, as they are near the place where human bones
were found in No. 2, they may have been derived from stratum
No. 2 and, as the writer has previously suggested,' may pertain
to the same individual as the bones found in No. 2.
The second discovery of human remains referred to stratum
No. 3 was made by Isaac M. Weills, who in April, 1916, obtained
a single human toe bone from the base of stratum No. 3 on the
north bank of the canal, 419 feet west of the bridge. A third dis-
covery of human skeletal material from this stratum was made
also by Mr. Weills, who in June, 1916, obtained a single human
tooth from the base of No. 3, on the north bank of the canal, 450
feet west of the bridge. All of the human skeletal material obtained
at Vero has been placed in the hands of Dr. A. Hrdlicka, who, it is
hoped, will discuss their relation to the modern races.
t Fla. State Geol. Surv., Eighth Annual Report, p. 142, October, 1916.
14 E. H. SELLARDS
In addition to the skeletal material a large number of artifacts
have been obtained from stratum No. 3. A small arrowhead was
taken from this stratum by Mr. Weills in April, 1916. A rather
large arrowhead was obtained from near the base of the deposit
by the writer in June, 1916. Two spalls have been found near
the contact line of strata Nos. 2 and 3, and four others have been
obtained in siftings from No. 3. From this deposit has been
taken about two dozen bone implements, as well as one wood
implement and one object, probably a section from an alligator
tooth, which apparently was used as an ornament. Broken pot-
tery is not uncommon in this stratum, about one hundred or more
pieces having been collected. Neither the pottery nor the bone
implements are restricted in their occurrence. On the contrary,
they are common to this horizon and are found at places where the
horizon reaches its maximum thickness and retains its covering of
marl rock, as well as at other places where the stratum is thinner
and has been cut into by the recent stream bed.
FOSSILS FROM STRATUM NO. I
Stratum No. 1, as previously noted, is a marine deposit in which
invertebrates are abundant and well preserved, the natural colora-
tion of the shells being in some degree retained. No vertebrate
fossils are known from this stratum at this locality, although from
shell marl, probably equivalent in age, at West Palm Beach the
writer has obtained the distal end of the humerus of a camel.
A collection of invertebrates from this horizon was made by the
writer in 1915 and submitted to Dr. T. W. Vaughan, in charge
of Coastal Plains investigations of the United States Geological
Survey. The mollusks of this collection have been identified
under Dr. Vaughan’s direction by Mr. Wendell C. Mansfield. Of
forty-two species definitely identified (17 gastropods and 25 pele-
cypods), all, according to notes kindly supplied by Mr. Mansfield,
are identical with the species believed to be now living. One
species only, an Arca, is regarded by Mr. Mansfield as intermediate
between the recent A. ponderosa and the probably extinct A.
limula. It appears, therefore, that the marine molluscan fauna of
FOSSIL HUMAN REMAINS AT VERO, FLORIDA 15
stratum No. 1, as well as the land and fresh-water molluscan fauna
of stratum No. 2, is essentially identical with the modern molluscan
fauna. In this respect the invertebrates are in very decided con-
trast to the vertebrates among which are many extinct species.
FOSSILS FROM STRATUM NO. 2
To the mammals previously listed from stratum No. 2? may
now be added the genus Mylodon, evidence of the presence of which
in stratum No. 2 was obtained in connection with the conference
of geologists and anthropologists held at Vero in October. Among
other important fossils added in November were about thirty
bones from the skeleton of the large extinct wolf, Canis ayersi.
These bones were found in the canal bank, from 7 to 10 feet west
of the place from which the skull and femur which served as the
type specimen of the species were found, and probably belong to
the same individual. From the north bank of the canal, 460 feet
west of the bridge, was obtained a practically complete skull of
the tapir, lacking only the lower jaw. From the south bank of
the canal at the railroad bridge was obtained about 44 feet of the
tusk of a proboscidian. ‘This tusk was found at the same place
and probably pertains to the mastodon, a part of the skull of which
had previously been secured. ‘The recovery of these fossils was
due chiefly to high water in the canal, following heavy rains at
the close of October. The water cleaned the banks of the canal,
thus facilitating both the examination of the section and the col-
lecting of fossils. To the birds, of which two species were pre-
viously known, a third species, represented by a humerus, may now
be added. ‘To the other fossils of this stratum—the plants, inver-
tebrates, fishes, batrachians, and reptiles—no species so far as
known are added in the new collections. The land and fresh-
water invertebrates, which include about twenty-eight identi-
fiable species, have been determined as previously reported by
Dr. Paul Bartsch and are found to be identical with the modern
species. The turtles have been identified by Dr. O. P. Hay and
have been found to include chiefly extinct species.
t Fla. Geol. Surv., Eighth Ann. Reft., p. 158.
16 E. H. SELLARDS
The following is a list of the mammals of stratum No. 2, which
fully establish the Pleistocene age of the deposit:
Didelphis virginiana Mammut americanum
Megalonyx jeffersonii Elephas columbi
Mylodon sp. Neofiber alleni
Chlamytherium septentrionalis Sylvilagus sp.
Dasypus sp. Sigmodon sp.
Equus leidyi Cryptotis floridana
Equus complicatus ) Blarina sp.
Equus littoralis Smilodon sp.
Tapirus sp. Hydrochoerus sp.
Odocoileus sp. Procyon lotor
Bison sp. Lutra canadensis
Peccary, indt. Canis ayersi
Camel, indt.
The Pleistocene fauna of this horizon is found at many places
in Florida. Some of the localities on or near the Gulf Coast from
which a typical representation of this fauna has been obtained are
Peace Creek, Sarasota Bay, the Caloosahatchee River, the With-
lacoochee River, and cave deposits at Ocala. Other localities on
or near the Atlantic Coast from which this fauna is known include
Daytona, Fellsmere, Palm Beach, Eau Gallie, and St. Augustine.
The mammalian species known from these different localities have
been listed by the writer in a paper recently published.
FOSSILS FROM STRATUM NO. 3
The fossils of stratum No. 3, which include plants, insects,
land and fresh-water mollusks, fishes, batrachians, reptiles, birds,
and mammals, have not been fully identified, and hence cannot be
discussed in detail at this time. Mr. E. W. Berry has kindly under-
taken the identification of the plants. The insects, of which only
a few have been obtained, have been submitted to Professor
H. F. Wickam. The land and fresh-water mollusks are few in
number, and presumably are identical with the modern. species,
since those of the older deposit, stratum No. 2, according to
Dr. Bartsch, are not separable from the modern forms. The fish
« “Fossil Vertebrates from Florida; A New Miocene Fauna; New Pliocene Species;
the Pleistocene Fauna,” Fla. Geol. Surv., Eighth Annual Report, pp. 77-119, Pls.
IO-14, 1916.
FOSSIL HUMAN REMAINS AT VERO, FLORIDA 17
remains, which are fragmentary, are in the hands of Dr. Charles
R. Eastman. The birds of this deposit are being studied by
Dr. R. W. Shufeldt.
The turtles from stratum No. 3 have been studied by Dr.
O. P. Hay, who finds that six species are present. Of these six
species three are extinct, one is sub-specifically different from the
modern, and two are apparently not separable from the modern
species. The mammals, the identification of which has been
approximately completed, are more abundant than the turtles.
The species of mammals recognized include the following:
Didelphis virginiana Scalopus sp.
Chlamytherium septentrionalis Vulpes sp.
Dasypus sp. Canis cf. latrans
Odocoileus sp. Procyon lotor ?
Neofiber alleni. Lutra canadensis
Sylvilagus sp. Lynx sp.
Sigmodon sp. Ursus, indt.
Neotoma sp.
The extinct genus of armadillo-like animals, Chlamytherium,
is represented by well-preserved, uneroded dermal scutes. The
armadillo, Dasypus, is likewise represented by dermal scutes. The
fox, which differs from the species at present known in Florida,
is represented by a part of the lower jaw containing two pre-
molar teeth, and by a single premolar tooth obtained from the
fresh-water marl rock on the south bank of the canal, 335 feet
west of the bridge. The rock at this place, as previously men-
tioned, has heretofore been placed in stratum No. 2, but at present
is regarded as probably equivalent to stratum No. 3. Canis cf.
latrans is represented by a part of the upper jaw containing the
carnassial tooth. The lynx is represented by a jaw and a tibia.
Parts of the teeth of Elephas columbi and of Mammut americanum
are by no means uncommon in this stratum. The tapir and horse
are also represented, although by fragmentary material.
Dr. R. W. Shufeldt has very considerately submitted an abstract
of his report on the fossil birds found at Vero, with permission to
insert it here in advance of the publication of the report as a whole.
The abstract includes all bird material obtained at Vero except a
18 E. H,. SELLARDS
stork, Jabiru weillst, previously described by the writer, and a left
humerus of a passerine bird related, according to Dr. Shufeldt, to
the meadow larks. ‘These two species and the first of the following
list, No. 7550, are from stratum No. 2 of the section. All other
birds of this list are from stratum No. 3.
REPORT ON FossIL BIRDS FROM VERO, FLORIDA, BY R. W. SHUFELDT
No. 7550. The right humerus, nearly perfect, of a teal. This bone I
have carefully compared with all the humeri of our smaller existing ducks of
various genera. It comes quite close to Querquedula discors, but belonged
to a different species of that genus. I propose to describe it as Querquedula
floridana.
No. 6934. This is the distal moiety of the right tibio-tarsus of a barn owl
(Tyto pratincola). The condyles are slightly chipped off posteriorly.
No. 6773. Distal half of the right tarso-metatarsus of a water bird;
probably a Larus, about the size of Larus atricilla. Whether this belonged to
the same specimen as the next (No. 6933) I cannot say.
No. 6933. Left carpo-metacarpus of some tern or gull; I am inclined to
believe, from the slight preponderance of characters, a gull of the genus Larus.
It was a considerably larger species than Larus atricilla, but comes close to it.
It was also a larger bird than Stirna maxema. I have compared it with the
corresponding bone in many species of terns and gulls. Apparently it does
not represent any of our existing forms. For this doubtless extinct species
I propose the name of Larus vero.
No. 7552. Humerus (right side), imperfect; head of bone not recovered.
Length of fragment 8.35 cm. No.6797. The shaft of an ulna of a large bird.
No. 6932. Piece of a long bone, the shaft (humerus?) of some large bird.
These three specimens too fragmentary for reference.
No. 7005 (in two pieces; smaller fragment not numbered). The left ulna
of Cathartes aura.
No. 6774. The distal two-thirds of the left tarso-metatarsus (imperfect),
of some heron (Ardea), larger than Nycticorax n. naevius. Not quite perfect
enough for exact reference.
No. 6932. The distal portion of the left tarso-metatarsus of Ardea herodias
(adult); the trochleae slightly abraded posteriorly.
No. 7554 (including three vertebrae). The two small vertebrae are from a
bird, and belong to the distal end of the cervical chain. It is hardly possible
to say to what kind of bird they belong, though they agree more or less with the
posterior cervical vertebrae of several average waders. The large elongate
vertebra is from the cervical region of another wader, a true heron of the genus
Ardea. I have compared it with the corresponding bone in the neck of all
our medium-sized waders, as herons, spoonbills, egrets, and many others, and
I find that, in all particulars, it comes nearest to the same vertebra in Herodias
FOSSIL HUMAN REMAINS AT VERO, FLORIDA 19
egretta. However, I would hardly feel justified in making a new species of this,
unless it was associated with other bones belonging to the same skeleton. I
would suggest, therefore, that it be set aside to await the discovery of additional
material from the same locality and the same excavation.
In this little lot there still remains the distal extremity of a small right
tarso-metatarsus, which is quite perfect as far as it goes. It belonged to some
sort of average-sized wading bird, perhaps after the heron order, or a near ally.
I have compared it with the corresponding part in some thirty skeletons of
existing birds; but, while it comes pretty close to some of them, it presents
departures of such a nature that it does not agree with any of them. I am not
prepared to describe it as coming from a new and extinct bird; but I would
suggest that it be set aside to await the discovery of additional material from
the place where it was found.
No. 7551. This is the distal portion of a right tibio-tarsus of a heron some-
what smaller than Ardea herodias, but specifically distinct from it. The con-
dyles are considerably abraded, but otherwise the specimen is perfect as far as
it goes. The intercondylar valley is narrower and shallower than we find it in
Ardea herodias, and the anterior tendinal groove in the fossil rapidly con-
tracts as it proceeds up the shaft, to become very narrow about 2 cm. above
the osseous tendinal bridge. This is not the case in Ardea herodias, wherein the
anterior surface of the shaft in that locality is flat, and barely exhibits any
tendinal groove. There are a few other points of difference, which, taken in
- connection with what is set forth above, inclines me to believe that this bone
belonged to a heron of the genus Ardea, of about the same size as the existing
Ardea herodias, but specifically distinct from it. For this apparently extinct
heron IJ here propose the name of Ardea sellardsi.
No. 7ooo. An imperfect distal third of the right tarso-metatarsus of a
large wader. Unfortunately, the trochleae are nearly all fractured off; still
the characters of this bone are so pronounced that I have no hesitation in
referring it to some species of Mycteria, and it probably belonged to a wood
ibis, Mycteria americana, with which I have compared it. So far as the frag-
ment goes, it does not seem to offer a sufficient number of characters to separate
it from that species.
It is thus found that stratum No. 3 of the section at Vero con-
tains extinct species of each of the three vertebrate classes—reptiles,
birds, and mammals. Since this deposit, No. 3, rests upon the
fossiliferous stratum, No. 2, it becomes necessary to inquire to what
extent these fossils may possibly have washed from the underlying
deposit. With regard to birds, those of No. 3 are more abundant
both in specimens and in species than are those of No. 2. Moreover,
the bird bones are fragile and would not withstand washing from
20 E. H. SELLARDS
one formation to another. That the extinct turtles do not repre-
sent inclusions from the older formation is evident by the fact that
practically complete carapaces are found, which in some instances
are so delicate as not to stand so much as turning over without being
broken. That the mammals referred to No. 3 are normal to that
deposit is indicated both by the abundance of the bones and by
their condition of preservation.
INTERPRETATION OF THE SECTION
It is desirable in this connection to consider the interpretation
of the section as a whole, especially as to whether or not an appre-
ciable period of time intervened between the deposition of divisions
r and 2 of the section, and also between divisions 2 and 3. Asa
rule, it is possible to recognize the dividing line between the marine
stratum, No. 1, and the fresh-water deposit, No. 2. At such places,
even though the marine shells are lacking in No. 1, there is a
change in the texture, and usually also in the color, of the sand.
Moreover, one finds rather commonly irregularities at the top of
No. 1, which are due to stream wash. Occasionally there are also
depressions or holes in the top of No. 1. Such holes, so far as
observed, as previously stated in this paper, contain muck and
decayed wood and sticks. However, notwithstanding this appar-
ently well-marked break, there are other places where the dividing
line between Nos. rx and 2 is evident neither by the change of
texture of the sand nor by any change in color. At such places
deposition appears to have been continuous from stratum No. 1
into stratum No. 2. On the other hand, there is at all places
evidence of the change from marine to fresh-water deposition.
Since the fossils of No. 1 are marine, while those of No. 2 are land
forms, there is little opportunity of connecting the two divisions
by means of the fauna. Aside from the one camel bone obtained
at West Palm Beach, no land animals are known from No. r of this
section.
In the preceding papers the writer has commented upon the
abrupt break which normally exists between strata Nos. 2 and 3,
a break which it seemed possible might be taken to indicate a con-
siderable interval of time. The later observations of this section,
FOSSIL HUMAN REMAINS AT VERO, FLORIDA 21
however, as previously stated in this paper, indicate that a part
of the section heretofore referred to No. 2 is possibly the equivalent
of No. 3. It seems not impossible, therefore, that the break
between divisions 2 and 3, which is so evident throughout a large
part of the section, may be due to local re-working by the stream of
its own bed in Pleistocene time, and that the deposit designated
as stratum No. 3 is itself a phase of stratum No. 2, being analogous
to the smaller deposits of muck and decayed wood found near the
base of No. 2, which are known to be inclusions within that stratum.
On the question of the interrelation of the three divisions of the
section, however, it will perhaps be necessary to await the accu-
mulation of further evidence, both stratigraphic and paleontologic.
RELATION OF THE HUMAN REMAINS TO THE ASSOCIATED FOSSILS
It will scarcely be maintained that the human remains and
artifacts obtained from stratum No. 3 are otherwise than normal
to that deposit. ‘Their abundance, their general distribution, and
their presence within and at the base of a stratified and undisturbed
deposit preclude any reasonable contention that they are other-
“ wise than contemporaneous with the associated materials of the
deposit. The study of the fossils of this stratum, although not yet
completed, has brought to light the presence of a considerable
number of extinct species which suggest the reference of the
deposit to the Pleistocene period.
Special interest is attached to the human remains and artifacts
from stratum No. 2, this being the oldest deposit from which
human material has been obtained. ‘This stratum is easily recog-
nized, and at the present time may be followed on both the north
and south banks of the canal through the whole section. The
vertebrate fauna by which the Pleistocene age of the deposit is
determined is also well represented in the collections that have
been made at this locality. That the human bones are fossils
normal to this stratum and contemporaneous with the associated
vertebrates is determined by their place in the formation, their
manner of occurrence, their intimate relation to the bones of
other animals, and the degree of mineralization of the bones.
The presence of flint spalls, and the probable presence of bone
22 E. H. SELLARDS
implements add support to the evidence obtained from the bones
themselves.
The place of the human bones in the formation affords a strong
argument for their contemporaneity with the associated fossils.
Those human bones found in the south bank of the canal, 330 feet
west of the bridge, lie beneath 18 inches of marl rock. The human
bones found in the same bank, 462 feet west of the bridge, lie
beneath 4 feet of stratified deposits consisting of alternating
layers of sand and muck, which could not have been dug through
and replaced without interrupting the continuity of the strata.
Moreover, the presence of the muck, as well as the conditions of
preservation of the plant remains, indicates that this locality has
been continuously moist since the materials of both Nos. 2 and 3
were deposited. Aside from the improbability of locating a grave
in a muck bed, it is probably impossible without special appliances
to dig a grave through an undrained muck bed on account of the
presence of ground-water. If it be suggested that the human
remains in stratum No. 2 represent a burial, it must be recognized
that the reference is not to a recent burial, but to a burial ante-
dating the deposition or existence of stratum No. 3 of the section,
and hence to an event that occurred probably within the Pleisto-
cene period of time. ‘There is, however, strong evidence that the
human remains in this deposit do not represent a burial by human
agency, but are fossils normal to the stratum, having been included
in the earth in the same way and at the same time that the other
bones were buried in the accumulating deposits.
The manner of occurrence of the human bones is entirely
similar to that of the other vertebrate fossils. Whole skeletons
are not found, and, indeed, complete bones are by no means com-
mon. On the contrary, the human bones as well as the bones of
the other animals are scattered, imperfectly preserved, and fre- —
quently broken. ‘The breaks in the bones are as a rule sharp-
edged, and it would seem that in the case both of the human and
of the other vertebrates the bones were more or less disturbed after
they had lost enough of the organic matter to become sufficiently
brittle to break as they were moved about by water before reaching
their final resting-place. An illustration of the way in which the
FOSSIL HUMAN REMAINS AT VERO, FLORIDA 23
bones were broken is afforded by a fragment from the shaft of
a human limb bone, No. 6958. This piece of bone is broken with
a sharp fracture at each end. The breaks are old and were made
at the time that the bone was imbedded in the sand. The intimate
association of the human and other fossils is difficult to explain
except upon the recognition of their contemporaneity. The
human astragalus and cuneiform were separated by a space of
about to inches, and between them at the same level as the
astragalus was the scapula of a deer. Bones from the skeletons
of other animals were near by in the same stratum and have been
described in the writer’s previous papers.
In the number of bones that have been obtained representing
a single individual there is observed no important difference
between the human and the other animals. The most nearly
complete skeleton that has been obtained is that of an alligator.
The next largest number of bones found, about thirty or thirty-
five in number, from the skeleton of a single individual are those
of the extinct wolf, Canis ayersi. Among the most perfectly pre-
served individual bones that have been obtained from the deposit
are those of the bird, Jabiru? weillsi, the femur of a horse, the
lower jaw of Chlamytherium, and part of the skull and tusk of a
proboscidian. Entirely surpassing any of the human bones in
perfection of preservation is the skull of the extinct wolf and the
skull of the tapir. The tapir skull, so far as the writer has been
able to learn, is the first approximately complete skull of this
animal that has been obtained from the Pleistocene of North
America.
The degree of mineralization of the human bones is identical
with that of the associated bones of the other animals, a fact that
has been brought out in the papers previously published by the
writer. The spall found in place in this stratum, as well as the
several other flints obtained from siftings, is totally unlike any
flint pebbles in the deposit. The reasonable explanation of the
presence of these spalls in this stratum is that they washed in from
the near-by land surface, together with the other materials of the
deposit. In other words, they pertain to the Pleistocene period
and were washed into the deposit at that time. The bone
24 E. H. SELLARDS
implements, although obtained from the siftings and hence not
seen in place, are, with little doubt, to be attributed to the same
source.
The presence of man in the Pleistocene of Europe has long been
known, and his assumed absence from the Pleistocene of America
is based entirely on negative evidence. How insecure as a basis
of argument in paleontology is negative evidence has been repeat-
edly demonstrated, and new groups with Old World affinities are
constantly being recovered from the North American formations.
A striking illustration is the eland obtained in 1913 by Gidley from
the Pleistocene of Maryland, the relationship of which is closer
to the modern eland of Africa than to any other known species, the
dispersion and migration of the group having probably occurred
_ during the Pleistocene period.t Another illustration is afforded
by the bears. Heretofore, it has been assumed that members of
the bear group were comparatively recent migrants to the New
World, but during the past summer representatives of the true
Ursidae were obtained practically simultaneously in Oregon? and
in Florida.3 | Numerous other illustrations might be given, and in
fact the rapidity with which new species are being obtained and
described is evidence of our heretofore imperfect knowledge of the
Pleistocene faunas. Man lived with and hunted Elephas primi-
genius in Europe, and it is not improbable that he may have
followed the spread of that species to America. ‘The evidence
obtained at this new locality in Florida, supplementing the less
positive evidence that has heretofore been available, affords proof
that man reached America at an early date and was present on
this Continent in association with a Pleistocene fauna.
«James Williams Gidley, ‘““An Extinct American Eland,” Smithsonian Miscel-
laneous Collections, LX, No. 27 (March, 1913).
2 John C. Merriam, Chester Stock, and Clarence L. Moody, ‘‘An American
Pliocene Bear,” Univ. of Cal. Publ., X, No. 7 (November 1, 1916).
3E. H. Sellards, “‘ Fossil Vertebrates from Florida,” Fla. Geol. Surv., Eighth Annual
Report (October, 1916).
INTERPRETATION OF THE FORMATIONS CONTAINING
HUMAN BONES AT VERO, FLORIDA
ROLLIN T. CHAMBERLIN
University of Chicago
The formations of the locality involved in the interpretation
of the age and relations of the fossil remains of man found near
Vero, Florida, have been clearly described in the preceding paper
by Dr. Sellards. The surface aspect of the region is plane and
flat, relieved slightly by low beach ridges, gently rising dunes, and
Fic. 1.—Rough sketch showing merely the general relations of the features dis-
cussed in the text.
shallow drainage flats, none of which are impressive features of
the landscape, though all contribute valuable criteria to the inter-
pretation. In its immediate bearings on the problem raised by
the occurrence of human remains mingled with extinct vertebrates,
the critical feature of this plain is the broad, shallow valley of
Van Valkenburg’s Creek, whose former course—now much obscured
by the recently dug drainage canal—is indicated in Figs. 1 and 2.
It was in the bottom deposits of this wide stream channel, or in
those of its predecessor, that the human bones in question were
found. The past workings of this stream, therefore, require the
25
ZS
20 ROLLIN T. CHAMBERLIN
closest scrutiny. But before attempting to interpret the history
of the stream, let us review briefly the geologic section at Vero,
though we have little occasion to depart from the careful descrip-
tion of Dr. Sellards. :
Beneath the stream deposits, as well as beneath the whole
region under consideration, the oldest beds exposed to view belong
to the Anastasia formation, a striking marine shell marl, often
known as ‘‘coquina’”’ rock. Composed almost entirely of shells,
he Contour Interval 5 feet
~ Scale
Ss 100 200 .
Feet
Fic. 2.—Detailed map of the locality where the human bones were found. The
canal and the resulting dump piles have done much to change the original topography.
The dotted area represents the flood plain of Van Valkenburg’s Creek as its appears
to have been just prior to the digging of the canal. The first human skeleton was
found in formation No. 2 at point marked M, the second at point NV, while human
relics were found in formation No. 3 at point O as well as also near VV.
in most cases but little fragmented, its identity is always clear, and
it thus affords a most excellent datum plane at the base of each
section studied. Itis Dr. Sellards’ formation No. 1, and is assigned
to the Pleistocene. In its upper portion it grades into beach sands
containing here and there lenses and streaks of shells, and these, in
turn, pass upward almost imperceptibly into fresh-water or wind-
blown sands. This formation underlies both the area of Van
FOSSIL HUMAN REMAINS AT VERO, FLORIDA 27,
Valkenburg’s Creek and the adjacent country, and thus far the
sections are the same, but above this horizon the channel section
and the upland, or country, section are totally unlike and will
be described separately.
THE SECTION OF THE CREEK BOTTOMS
This is the section described by Dr. Sellards. It is now well
exposed in the walls of the drainage canal which, for several hundred
Fic. 3.—The drainage canal, looking southwest from the Florida East Coast
Railroad bridge to the spillway.
yards, cuts through the deposits of the old creek bottoms. Resting
upon the eroded surface of formation No. 1, sometimes lying
directly upon the coquina rock, sometimes resting upon the beach
sand, or shore phase of No. 1, is Dr. Sellard’s formation No. 2.
It is, as he has described it, a cross-bedded, river-washed sand,
partly white, partly stained brown by organic matter, and contain-
ing partially decayed wood and muck. At the top, in places,
there is a fresh-water, clayey marl. This formation contains
human bones essentially in sztu, beyond reasonable doubt, together
28 ROLLIN T. CHAMBERLIN
with the scattered bones of many extinct vertebrates, as main-
tained in the previous paper. It is, therefore, the critical formation
of the section, and upon its age and mode of origin the case of
Pleistocene man stands or falls.
Above formation No. 2, and, at most points, sharply separated
from it by a clear-cut line of erosion, is an alluvial deposit, forma-
tion No. 3 of Sellards. ‘This is composed of swamp muck in many
layers, interstratified with layers and lenses of coarse sand. Its
top is the present flood-plain surface of Van Valkenburg’s Creek.
INTERPRETATION OF THE CREEK SECTION
Following the deposition of the marine coquina, a withdrawal
of the sea gradually brought this region into the horizon of terres-
trial action. In the transition, beach conditions prevailed, result-
ing in sandy deposits, partly marine, partly terrestrial. These
finally gave way to eolian sands. An interval of unknown duration
followed. At some later time a stream occupying essentially the
same course as that which, just prior to the construction of the
canal, was followed by Van Valkenburg’s Creek excavated a
channel which, in some places, was cut through into the coquina.
The notable width of this channel in proportion to its very shallow
depth—which was limited by sea-level—suggests that erosive
action was in progress for a considerable time at least. But as
Dr. Sellards has remarked, tidal scour may have been an acces-
sory factor in the development of the breadth of the channel.
There are today in the strip of coast between Sebastian and Eau
Gallie several such broad, shallow channels up which the tide runs.
But if tidal scour is appealed to, it must be interpreted so as to
be consistent with the fact that there were deposited in the Vero
channel, not only muck and washed sands, but also human bones,
together with many scattered bones from a variety of extinct
vertebrates. The interpretation must also be in harmony with the
specific fact that the human bones were found to be notably less
scattered and fragmentary than the bones of the extinct verte-
brates.
Some erosion of the surface of this No. 2 formation seems
clearly to have occurred, since its upper surface is a sharp line,
FOSSIL HUMAN REMAINS AT VERO, FLORIDA 29
and this is most naturally interpreted as implying change of atti-
tude or of relations, as well as an erosion interval. After such
erosion it was covered by the alluvial deposit, No. 3. Because of
the large proportion of muck, and the extremely rapid shifting
from muck-accumulating to sand-depositing conditions, as revealed
by the many alternating layers and lenses of sand and muck, this
Fic. 4.—The creek section freshly exposed by cutting back into the south bank
of the drainage canal at a point about 440 feet southwest of the railroad bridge:
I represents the ‘‘coquina”’ rock grading upward into light-colored sands; 2 is
Sellards’ formation No. 2, which is here sand largely stained dark brown; note that it
fills a trench cut in No. 1; 3 is Sellards’ formation No. 3, consisting of alternating
muck and coarse sands. Large log projects conspicuously on the right.
is interpreted as a flood-plain deposit. Its upper surface consti-
tuted the flood-plain of Van Valkenburg’s Creek prior to 1913,
when the drainage canal was dug.
THE UPLAND OR COUNTRY SECTION
Downstream.—Though the surrounding country is not elevated,
the general section outside the valley of the creek may be desig-
nated the upland, or country, section. Downstream from the
30 ROLLIN T. CHAMBERLIN
locality where the human bones were found the country section
adjacent to Van Valkenburg’s Creek commences with the coquina
rock at the bottom, just as does the section of the creek bottom.
The coquina beds are followed by 2-5 feet of variously tinted sands.
An orange-brown, ferruginous sand is a very persistent phase.
These sands become dark-brown to blackish at the top, but are not
very firmly indurated. At most points this sandy deposit forms
the present surface, but in some isolated areas it is capped by a
pondlike deposit of drab-colored, clayey, fresh-water marl. As
this is being utilized for road material, the clay marl areas have
been opened up to view and their extent is well known.
The south bank of the canal one-third of a mile east of the
Florida East Coast Railroad bridge gives the following section:
0) Drab-colored; clayey, fresh-water marley jee eetansiee esis) ei aje- sore 2) 5B
n) Dark-brownish, mottled sand, lighter colored below, getting darker
above and at the top showing nearly black material, indicating an old
The tract represented by this section lies nearer the coast than
' the locality where the human bones were found in the stream
deposits, and hence has less specific bearing on our problem than
the upland section of the tract adjacent to the creek above the
critical locality. a hi ;
Upstream.—The upstream section was found to be somewhat
different from the coastward section as given above. Approxi-
mately 200 feet southwest (upstream) from the point where the
human relics were discovered the waters of the drainage canal
pass over a spillway and drop about 9 feet. This spillway is west
of the junction of the two tributary branches of Van Valkenburg’s
Creek and lies outside the creek valley. For the first half-mile
west of the spillway the canal has been cut through the following
succession of beds:
d) Pure-white, coarse-grained, wind-blown quartz sand............. 4-7 it.
c) Soft, spongy, peaty layer, containing many partially decayed roots;
in: places; absent 4 ei sW4iy Pe aa sled aE Co OA REN ea Sen cs NU Sana o-6 in.
b) Dark-brown to true-black, firmly indurated sand or sandstone;
cemented by ferric hydroxide and organic matter, but the color of
iron staining is largely obscured by the organic black........... 2-4 ft.
FOSSIL HUMAN REMAINS AT VERO, FLORIDA at
a) Brown sand gradually losing its dark stain and passing downward
into a reddish-brown sand stained by iron oxide, and finally grading
into a buff sand below, which is of finer grain than that above, and
MT AVA OOSSIO LYM COMIATINE tyeiite nein Lok Lk avai E NTC iil 3-4. tt.
There is no sharp division between (a) and (6).
For at least another half-mile west the section changes in no
essential feature except that the wind-blown sand (d) gradually
Fic. 5.—The creek section on south bank of drainage canal, 465-70 feet from the
bridge. This face, now dug back many feet from the original drainage canal bank,
is approximately the spot where the second human skeleton was found in formation
No. 2 (marked N on map, Fig. 2). Formations Nos. 2 and 3 displayed.
thins until, at one mile from the spillway, there are barely two
feet of it. The coquina rock does not appear above the water-
level in the canal during the first mile west of the spillway, though
it is said to reappear some distance farther west.
INTERPRETATION OF THE UPLAND SECTION
Following the deposition of the marine coquina beds, semi-
marine, semi-terrestrial beach sands, and, later, eolian sands
accumulated to a thickness of perhaps 6-8 feet over much of the
32 ROLLIN T. CHAMBERLIN
country immediately northwest of Vero. In spots the thickness
was less, but at other points, no doubt, dunes as well as beach
ridges made its total more than this. A widespread peaty layer
at the top of this formation strongly suggests that, following the
deposition of these sands, bog conditions existed for a period in
the area west of the spillway. The upper two or three feet of the
sands of layer 6 are very firmly cemented by iron oxide, and are
Fic. 6.—Section of uplands, exposed in north bank of drainage canal, one-third
mile southwest of spillway: B represents layer b, consisting of indurated black sand
or sandstone, capped in the left half of the picture by peat (layer c); Dis layer d, white
eolian sands, resting upon the Pleistocene bog surface. Above is the material exca-
vated in making the canal.
deeply stained by organic matter, implying that this horizon con-
stituted the subsurface for a long time. Iron oxide cement is, as
is well known, common in bog deposits. A reason for an extensive
wet area, or bog, to the west of the spillway is readily found in
a broad beach ridge near the spillway which interfered with the
drainage of the tract lying west (see Fig. 1).
The present coast, in a way, serves as an example of similar
relations. The east coast of Florida is flanked by a barrier sand
FOSSIL HUMAN REMAINS AT VERO, FLORIDA 33
ridge for 300 miles; for over 100 miles the barrier incloses a strip
of water between it and the mainland, known as the Indian River,
though it is really a salt-water sound. Paralleling the present
coastal barrier and the Indian River behind it is an older
barrier ridge which crosses the canal near the spillway and runs
for many miles both north and south of Vero. To the west
of it, before the drainage canal was dug, the region was frequently
Fic. 7.—The present upland country southwest of the spillway. The area of
the Pleistocene bog. Drainage canal in foreground with tributary canal in middle
distance. Lumps of the black sandstone conspicuous upon dump piles of both canals.
under water after storms, according to testimony, and in earlier
times it presumably was more continuously marshy, since it
more nearly approached the present condition of the Indian
River.
But with uplift, or withdrawal of the sea, the marsh was grad-
ually drained, and a thin covering of wind-blown white sand
drifted over the old bog surface, burying it to a depth of several
feet. This wind-blown sand forms layer d and constitutes the
present upland surface.
34 ROLLIN T. CHAMBERLIN
CORRELATION OF CREEK-BOTTOM SECTION WITH UPLAND SECTION
For the complete history of the district it is necessary to cor-
relate the creek-bottom section with the upland, or country,
section. The coquina is common to both and serves as a base of
reference. If we turn to the upland section as it is developed just
west of the junction of the two forks of Van Valkenburg’s Creek,
we observe that the most striking feature there shown is the almost
perfectly black, indurated sand bed, or sandstone, which forms
a persistent layer, in places capped by peat, beneath the surficial
wind-blown sands. If the creek deposits were younger than
the induration of this sandstone, evidence of such relative age might
well be found in the incorporation of derivatives from the black
sandstone in the creek deposit, for, if the age and induration were
considerable, the sandstone should have been of sufficient hardness
to supply the two forks of the creek with pebbles and cobbles of
this very easily recognizable material. *Now an inspection of the
freshly cleaned face of Dr. Sellards’ formation No. 2—the critical
formation—reveals the presence in it of many small pebbles, and
not a few round ‘‘cannon balls,” of this black sandstone (see Fig.
8). The latter range up to 5 inches in diameter. They are not
confined to any one layer, but are scattered through No. 2 forma-
tion from top to bottom. Thus the formation in which the human
bones and the extinct vertebrate remains were found also contains -
numerous pebbles and cobbles of black sandstone from the older
formation! ‘The stream which deposited formation No. 2 formed
these pebbles of black sandstone by erosive action on stratum b of
the upland section, through which both the north fork and the
south fork of Van Valkenburg’s Creek have obviously cut their
stream channels. This black layer underlies apparently all the
country immediately to the west of the spillway; it is a continuous,
persistent layer; it was traced im situ to within 150 feet of where
the human remains were found, and with further digging it could
probably be traced still nearer. There is no other known source
for the pebbles of black sandstone. The conclusion seems, there-
fore, inevitable that Dr. Sellards’ formation No. 2 is younger than
stratum b of the upland section—the old bog surface upon which the
peat accumulated. Furthermore, it would seem to be considerably
FOSSIL HUMAN REMAINS AT VERO, FLORIDA 35
younger, inasmuch as the old bog-covered sands had, since their
formation, endured for a time sufficient to permit their upper por-
tion to become firmly cemented by iron oxide into a fairly indurated
sandstone.
INTERPRETED HISTORY OF THE BONES
As is well described in Dr. Sellards’ paper, human bones were
found in situ in formation No. 2, in close association with scattered
Fic. 8.—The creek section on south bank of drainage canal, about 460 feet south-
west from railroad bridge. Many small pebbles (size of marbles) of black sandstone
here form a band in the lower portion of formation No. 2.
bones and fragments of bones from a great variety of extinct mam-
mals, including the Columbian elephant, mastodon, saber-tooth
tiger, tapir, armadillo, sloth, bison, camel, horse, etc. This verte-
brate fauna, according to Dr. O. P. Hay,’ would seem to represent
the early Pleistocene. But it should be noted again that, while
the human bones make up quite a part of two skeletons, the bones
of the extinct vertebrates are fragmentary and extremely scat-
tered. This fact, that the remains of the ancient vertebrates are
t Personal statement on the ground.
36 ROLLIN T. CHAMBERLIN
very fragmentary, by itself suggests that they have been disturbed
and transported to a greater or less extent. On the contrary, the
human bones, so much less scattered as to indicate that they
belong to distinct individual skeletons, imply that they have
suffered much less disturbance.
The history of the bones seems to unravel as follows: For along
time during the Pleistocene there existed a marshy area of con-
siderable extent immediately to the west of the present location of
the spillway. Peat accumulated in the bog, forming layer c of the
upland section. In the course of its growth, various animal
remains of the time became incorporated. The large vertebrates
were no doubt often mired, and left their bones in the bog. This is
a familiar process.
During and following the accumulation of the bog and the
bones, the upper portion of the sandy formation that lies beneath
the peat (stratum 6) became indurated to sandstone by the infil-
tration of iron oxide. At the same time it was stained black by
the decomposition products of the decaying organic matter that
lay over it.
With further passage of time the large land vertebrates, one
after another, became extinct in the region. In the course of time
also man appeared in Florida. At some time subsequent to the
growth of the bog, probably as the result of a slight uplift, Van
Valkenburg’s Creek and its branches, or their antecedents, cut
channels across the beach ridge and into the peat deposit, the two
forks apparently following essentially the courses which they hold
today. The drainage lines thus established cut, not only into the
Pleistocene bog deposits, but into the sands beneath. With the
draining of the bog and the adjacent land the movement of sand
by wind action was perhaps facilitated, and the dune formation
(layer d) which covers the old bog surface west of the spillway may
have been formed during the drainage stage. It is not incon-
sistent, however, to suppose that it was formed before the drainage
was established.
Later came the stages of partial filling of the creek channel.
The first stage was occupied by the deposition of formation No. 2
of Sellards. The material for this formation was derived from the
FOSSIL HUMAN REMAINS AT VERO, FLORIDA 3a
upland section drained by the creek, not a little of it from layer 6,
as shown by the pebbles of black sandstone. The whiter portion
of the sands of the stream deposit may have come from either layer
a or from layer d of the upland section. To layers 6 and ¢ are
assigned the bones of the extinct vertebrates together with the
pebbles and cobbles of black sandstone. While this deposition of
formation No. 2 was in progress, the human bones are believed to
Fic. 9.—The present dry channel of Van Valkenburg’s Creek just beyond the
reaches of the drainage canal.
have received their first and only burial in connection with the
stream deposit. ‘The human bones should thus naturally be less
scattered than the fragments of the mammals which had been shifted
from their original location into the stream channel.
Following the deposition of formation No. 2 there was a period
of erosion, either in the form of the ordinary scour-and-fill process
of streams or as a result of ordinary subaerial denudation. A
change in the stream conditions following the erosion stage caused
a very heterogeneous alluvial flood-plain deposit (Sellards’ forma-
tion No. 3) to be laid upon the irregular surface of formation No. 2.
38 ROLLIN T. CHAMBERLIN
Muck accumulated in alternation with layers of sand, as recurring
floods from heavy tropical storms carried coarse material before
them.
To this formation No. 3—just as in the case of formation No. 2
—the upland bog area contributed many bones of extinct verte-
brates as well as pebbles of black sandstone. At the same time
human bones, pottery, and bits of flint (which does not outcrop in
the region) were mingled with the flood-plain deposit, more or less
directly, it would appear, as the result of human activity. There
thus again came to be close assemblage of all this varied material in
this formation, just as there had previously been in formation No. 2.
These conditions are interpreted as having been continued with
little change (except on the human side) till the present, for pebbles
of black sandstone and bones of extinct vertebrates are found i
the deposit of the present creek bed into which formation No. 3 merges.
Two sets of evidence developed by Dr. Sellards need to be
explained if we are to accept the sequence of events above outlined.
Chemical analyses are cited as showing that the fossil human bones
from No. 2 are quite as well mineralized as are the associated bones
of the Pleistocene animals. Compared with a bone from an
Indian mound near Vero, the chief difference is that the bones from
No. 2 (human and other) have lost from 6 to 8 per cent of moisture
and from g to 11 per cent of volatile matter. The loss of these
easily eliminated constituents caused a proportionate increase in
the percentage of calcium and phosphoric acid. But there was,
in addition, an actual infiltration of silica, etc., from 0.4 to 2.9 per
cent, and of iron and aluminum oxides from 0.6 to 3.5 per cent.
While indicative of considerable age, it must be admitted that we
do not know how rapidly bones are thus altered in sandy river beds’
when the adjacent sands contain abundant iron oxide.
A carapace of the turtle, Terrapene innoxia Hay, taken from
formation No. 2 complete, though it was very fragile at the time of
discovery, and the skull of a large wolf, Canis ayersi Sellards, are
taken as evidence that the bones of the vertebrates were not trans-
ported from some other point by the creek. The turtle carapace
was too fragile in the fossilized condition in which it was found to
admit of stream transportation, though perhaps it could have
FOSSIL HUMAN REMAINS AT VERO, FLORIDA 39
endured transportation before fossilization. But if the interpre-
tation of the history of the region be as outlined above, it would
not seem unreasonable to suppose (in case it be definitely estab-
lished that these species have been extinct in the region since the
close of the Pleistocene) that the turtle carapace and the wolf
skull, and other similar parts, had been subjected to a minimum of
transportation wear because originally buried in the upland forma-
tion close to the spot where they were found, and that they were
carried into the channel fill by the caving of the river bank, or some
similar operation involving little wear. In no case was the trans-
portation great. The other bones found in No. 2 and No. 3, in the
opinion of the writer, give as much evidence of wear and polishing
as would be expected of bones that were washed only short dis-
tances (from the upland bog to the places in the channel where
they were found) by the flood stages of the creek.
Formation No. 3, therefore, seems to the writer to be very
recent geologically, as it is the flood-plain alluvium of the present
Van Valkenburg’s Creek. The age of formation No. 2 can be
determined less positively. It is simply older than No. 3 and
younger than the Pleistocene bog deposits that lie west of the spill-
way, but it is the opinion of the writer that it is much nearer in
age to No. 3 than it is to the Pleistocene bog accumulation and
associated deposits which originally housed the old mammalian
bones.
ON REPORTED PLEISTOCENE HUMAN REMAINS AT
VERO, FLORIDA
THOMAS WAYLAND VAUGHAN!
United States Geological Survey
Topographic relations.—Vero, a village on the Florida East
Coast Railway, 228 miles south of Jacksonville, in St. Lucie
County,’ is situated on the surface of the Pensacola terrace, the
lowest and youngest of the three Pleistocene terraces recognized
in Florida by Matson,’ and is about one mile west of the western
shore of Indian River, between which and the Atlantic Ocean lies
the great barrier beach of east Florida. The terrace plain presents
the physiographic aspect of early youth, as it is almost flat
and is only slightly trenched by rather indefinite drainage courses.
Its surface stands between 1o and 15 feet above sea-level, ex-
cept along an elevated barrier beach which lies some 600 or 700
feet west of the railroad, where the altitude may be as much as
to to 15 feet higher. The human remains were found in a slight
depression along a drainage course across the terrace surface at
localities about half a mile north of Vero and between 330 and 580
feet west of the railroad, and were exposed as a result of the
excavation of the Indian River Farms Company drainage canal.
Geologic relations.—Dr. Sellards has in three papers‘ presented
detailed descriptions of the geologic section exposed along the
« Published by permission of the Director of the United States Geological Survey.
2 See the map of State of Florida, scale 1/500,000, issued by the United States Geo-
logical Survey in 1916, and the United States Coast and Geodetic Survey Chart
No. 163.
3“Geology and Ground Waters of Florida,’ U.S. Geol. Survey, Water-Supply
Paper 319, pp. 31-35, Pl. 5, 1913.
4“Discovery of Fossil Human Remains in Florida in Association with Extinct
Vertebrates,” Amer. Jour. Sci., XLII (July, 1916), 1-18; ‘‘Human Remains and
Associated Fossils from the Pleistocene of Florida,’ Eighth Ann. Rept. Florida Geol.
Survey, 1916, pp. 122-60, Pls. 15-31; ‘‘Human Remains from the Pleistocene of
Florida,’ Science, N.S., XLIV (1916), 615-17.
40
FOSSIL HUMAN REMAINS AT VERO, FLORIDA 41
canal from the Florida East Coast Railway on the east to a point
about 580 feet westward from it. The stratigraphic succession
may be summarized as follows: (1) The lowest observed bed is an
arenaceous shell marl of Pleistocene age, the exposed thickness of
which is from 2 to 6 feet. (2) Unconformably above bed No. 1
are sands, some muck, and marl, having a combined thickness
ranging up to as muchas 5 or 6 feet. This formation was deposited
in fresh water and contains numerous species of vertebrates, which
clearly indicate its Pleistocene age, and shells of about 30 species
of land and fresh-water mollusks. The discovery of a locality at
which so many species of extinct vertebrates are represented is of
much geologic interest and importance. Within the sands human
remains were found at two places, according to Dr. Sellards.
(3) Overlying No. 2 is a deposit of muck, tree trunks, and other
vegetable matter, in which are stringers of sand, in places con-
taining marine shells, perhaps derived from bed No. 1 by erosion
farther upstream. This deposit was accumulated in a shallow,
relatively wide, channel eroded in No. 2, and has a thickness of 3
feet 6 inches in the middle of the channel, but it is much thinner
on the channel sides. Whether its geologic age is Pleistocene or
Recent has not been positively determined. Dr. Sellards reports
human bones from near the base of this bed and from sands which
lie at its base along the contact with No. 2."
Criteria for determining the geologic age of the human remains.—
Previous investigations having shown that human artifacts may,
by many agencies, be carried below the surface of the ground and
become imbedded in unconsolidated deposits, and as it is well
known that human bones may have been either naturally or arti-
ficially buried, the occurrence of artifacts and human bones in
association with Pleistocene fossils does not prove the Pleistocene
age of man. It seems to me that the only indisputable geologic
proof of the Pleistocene age of man must consist in finding a con-
tinuous undisturbed bed or layer of demonstrable Pleistocene age
above the human remains (artifacts or bones) whose age is under
investigation. The relative dissociation and the significance of
t Fighth Ann. Rept. Florida Geol. Survey, 1916, pp. 140-42, Pl. 17, Fig. 1, text
Fig. 14.
42 THOMAS WAYLAND VAUGHAN
the mineralization of the vertebrate (including the human) bones
is discussed by others.
Conclusions.—As bed No. 3 may be of recent geologic age, the
presence of human bones in it does not now need special consider-
ation. With regard to the remains in bed No. 2 it will be said
that as intrusion into it may have been accomplished either by
natural or by artificial processes subsequent to its deposition; the
presence of the human remains in it, in my opinion, is not definite
proof of their Pleistocene age. However, should it be postively
shown that in bed No 3 Pleistocene fossils occur in place above
the human remains, showing that subsequent to the death of the
individual represented by these remains Pleistocene species belonging
to other groups of organisms lived and died, the evidence in favor
of the Pleistocene age of the human remains would be conclusive.
On the other hand, should it be proved that bed No. 3 is of Recent
age, the human remains might be of either Pleistocene or Recent
age, and it is doubtful if positive criteria for determining their age
will be available unless the needed information is furnished by the
human bones themselves. As the accurate determination of the
geologic age of bed No. 3, especially that part of it perpendicularly
above the human remains, seems to me to be critical, it is my
opinion that, for the present, judgment should be suspended.
¥
7
PRELIMINARY REPORT ON FINDS OF SUPPOSEDLY
ANCIENT HUMAN REMAINS AT VERO, FLORIDA
ALES HRDLICKA
United States National Museum, Washington, D.C.
On the kind invitation of Dr. E. H. Sellards, state geologist of
Florida, and as his guest, the writer in the latter part of October,
1916, spent four days at Vero, Florida, where his time was devoted
to the study of the site from which certain human bones described
by Dr. Sellards were obtained, and to a preliminary examination of
the bones themselves.
Generous assistance in this work was rendered by Dr. Sellards
and his associate, Mr. Gunter, as well as by the two local gentlemen
most directly interested in these finds, namely, Messrs. Ayers
and Weills, to whom the writer wishes to express his grateful
acknowledgments.
On arriving at Vero the writer engaged workmen and with their
aid made a clean exposure about 160 feet in length of the geological
deposits in close proximity to the spots where the human bones
had been discovered. ‘This afforded a comprehensive and enlighten-
ing view of all the formations involved.
The two human skeletons had been found in the south bank of a
recently excavated drainage canal. They occurred one in fairly
close proximity to, and the other within the broad shallow bed of,
a small fresh-water stream, now drained by a lateral cut from the
canal. The former lay in dark and somewhat indurated sands,
layer No. 2 of Sellards, the latter for the most part at the base of
layer No. 3, the muck deposit of the stream bed, and “between this
and the next older stratum” (Sellards). A few smaller bones which
probably belonged to the second skeleton were found at about the
same level and at a short distance from the rest of the remains in a
small elevation of the irregularly eroded upper surface of the lower
sandy layer No. 2.
43
AA ALES HRDLICKA
The first skeleton lay at the depth of two and a half feet, the
second at the depth of from two to possibly three and a half
feet from the surface.t. The first was found accidentally and taken
out by Messrs. Ayers and Weills, before Dr. Sellards was notified,
and before any great importance was attached to the find. The
character of the deposits above it was not especially noticed, but
there is no reason for supposing that they differed from those in the
neighborhood, where layer No. 2 is seen to be overlain by a stratum
of similar, but somewhat lighter, sandy deposits covered by a layer
of marl. This marl ranges at this point from about 5 to 9 inches
in thickness, and when freshly exposed is of the consistency of
fresh mortar, but on exposure hardens to fairly solid rock. With
some wind-blown white sand and oie. material it forms the
surface of the ground.
The second skeleton lay, according to vt obtainable information,
in some loose white sand and vegetable matter at the base of the
muck layer, No. 3, of the stream bed. Above, up to the surface,
there was only muck with irregular sandy patches. In a vertical
cut these localized deposits or patches give the muck an appear-
ance of unconnected irregular lamination, but there are no actual
strata.
Skeleton No. I is that of a woman, possibly sub-adult. Skele-
ton No. II is that of a man, an adult of somewhat advanced years.
The bones of the former, according to Mr. Ayers, who discovered
and extracted them, “were all close together, the whole layer not
being over one and one-half feet in width. They were not scattered
at all, nor piled up.” The various parts lay side by side or next to
one another in about the position they would occupy in the body.
The bones of skeleton No. II were dissociated, though lying within
an ellipse apparently about 7 feet in length, not counting the two
bones and two or three fragments found in the upper part of
layer No. 2, about 6 feet away. As some of the bones of the skeleton
tumbled out of the bank before the rest were removed, only a
smaller portion of the parts representing the skeleton were examined
1 In Dr. Sellards’ report on the find, in the 8th Ann. Rep. of the Fla. St. Geol. Survey,
p. 142, the depth is given as 4 feet, which is evidently an error; the depth indicated
in Dr. Sellards’ illustrations, especially that on p. 141, is less than this.
FOSSIL HUMAN REMAINS AT VERO, FLORIDA 45
im situ and their exact association must remain in a large measure
uncertain. ‘The skeleton lay in an inclined plane. The bones show
no trace of washing or weathering. The majority of them are
broken, but many of the breaks are sharp and evidently fresh,
Fic. 1.—Top view of skull of skeleton No. II, from the base of muck bed (layer
No. 3), south bank of the drainage canal, Vero.
c=clay; portion of frontal bleached by exposure.
dating probably from the time when parts of the skeleton were
exposed in the bank or tumbled out of it.
Bones of three other individuals are found in the collection
made by Dr. Sellards’ party. They are a juvenile or a young
adult incisor tooth from layer No. 3, in the vicinity of skeleton
No. II; a tooth of a young child from stratum 3 on the opposite
46 ALES HRDLICKA
or north side of the canal, and a toe bone of an adult, also from
the north side of the canal.
In the muck layer on the south side, in the base of which skeleton
No. II occurred, there were found, according to Dr. Sellards, “an
abundance of pottery, many bone implements, arrowheads, and
other small flints.”
Speaking further on this point, Dr. Sellards says (p. 143):
A considerable amount of broken pottery is found in this horizon, par-
ticularly at the locality on the south bank 450 to 475 feet [bones of skeleton
No. II were located from about 460 to 473 feet] west of the bridge. Bone
implements are also numerous and were made evidently to serve a diversity
of purposes. Well-worked flint arrowheads are found also, as well as occasional
spalls from the manufacture of flints. The pottery, flints, and bone imple-
ments, however, are not confined to this locality on the south bank, but are
found also in the same horizon on the opposite side of the canal.
A few small flints and two bone implements were found in
stratum No. 2 (p. 140). The flint of the several chips and imple-
ments, which must have been brought from a considerable distance,
is quite similar in the two deposits; and the bone implements of
the two sections seem identical in character.
The portion of the muck of the stream bed on the south side of
the canal nearest where the bones of skeleton No. II were dis-
covered was found to be a moderately compressed, wet mass of
leaves and other detritus. Many of the leaves, though generally
imperfect, were still so pliable that they could be unfolded and
straightened out, and were still fairly elastic. In this muck are
trunks of trees and branches or roots, partly in a fair state of
preservation, partly softened or rotted.
During the clearing work carried on by the writer, fossil animal
bones were found to be fairly numerous in layer No. 2, beneath the
muck of the stream bed. There were uncovered possibly several
hundred specimens of this nature. They were isolated, small
and large fragments, some apparently waterworn, with a few
individual bones, and parts of turtle shells. The largest individual
specimen was the tooth of a large herbivore. Two or three frag-
mentary fossilized bones were also obtained from a sandy band in
the lowest portion of the muck deposit.
FOSSIL HUMAN REMAINS AT VERO, FLORIDA 47
The foregoing comprises in brief the writer’s personal observa-
tions at Vero, with the exception of those on the human bones
themselves. After a careful’ weighing of the facts, both on the
spot and afterward, he regrets that he cannot agree with the con-
clusions reached by Dr. Sellards as to their antiquity. It seems to
him that there is another possible and more likely explanation of
their occurrence in the deposits than that which would make
Fic. 2.—Right side of skull belonging to skeleton II (frontal bone, light from
exposure, on the right, occiput on the left).
them contemporaneous with the various fossil animals the remains
of which are found in the same layers, and some of which may date
from the middle or even early Pleistocene.
A relatively small amount of work brought to light the remains
of five human individuals—a small child, an adolescent or young
adult, a young woman, and two adult men. In the vicinity of
these occurred a quantity of pottery fragments, resembling closely
the usual Florida variety, bone implements, and stone imple-
ments with chips, and all in proximity to, or in, the bed of a
48 ALES HRDLICKA
fresh-water stream. To the anthropologist the various finds.
strongly suggest an ordinary “station,” or inhabited site, with
burials of probably prehistoric, but not necessarily very ancient,
man, whose culture horizon corresponded to that of the ordinary
American aborigines of the eastern and southeastern states.
The two human skeletons occurred at nearly the same depth,
which would be about that of a common Indian burial. ‘The bones
of the one were in close and natural association; those of the other,
buried in or just below the unstable muck, though dissociated, yet
remained fairly well aggregated, preserving some original relations.
The condition of these remains, contrasted with that of the animal
fossils with which they were associated, is instructive. ‘The num-
ber of individual fossil animal specimens recovered by the local
explorers, Dr. Sellard’s party, and the visiting scientists would
doubtless reach several thousands, and they were with a few
exceptions isolated bones or teeth or mere fragments, many of
which were hardly worth collecting.
The occurrence of isolated fossil animal bones or fragments in
contact with, or even above, the human skeleton would have no
significance. In digging a grave the earth thrown out might well
contain fossils even of considerable size, which, after the body was
introduced, would be thrown in about or above it.
The apparently undisturbed condition of the partial and
irregular sandy layers which occur in the muck where skeleton
No. II was discovered could hardly be regarded as sufficient proof
that the bones were not introduced from above. The muck and
sand thrown in over a body would tend in the course of time so
completely to assume the appearance and characteristics of the
original deposits that distinction between the two would be quite
impossible. Very good examples of restratification and striation
are seen at Vero in the accumulations thrown out from the canal
by the dredges.
The human bones are considerably ‘‘fossilized.” But they are
not fossilized equally in the two skeletons,‘ nor even in the different
parts of one and the same skeleton. The mineralization also is not
« The considerably smaller female astragalus weighs 26 grams, the much larger
male bone but 20.7 grams.
FOSSIL HUMAN REMAINS AT VERO, FLORIDA 49
quite like that of the animal bones from the same deposits, though
the approach, especially in parts of skeleton No. II, is close. Even
if they were identical, however, in this respect, the fact could not be
taken as a gauge of their contemporaneity with the animal bones.
Mineralization is a chemical-mechanical process, which runs its
course slowly or rapidly, according to circumstances. Under
similar conditions two bones, ages apart, would ‘‘fossilize”’ in a
similar manner; but one of the bones would have completed the
process long before the other. The writer has dealt with this
subject in his report on “Ancient Man in North and South Amer-
ica.”? In the corresponding work on North America will also be
found described examples of human bones, petrified in different
ways, from the west coast of Florida. One of the skeletons from
that locality, in the possession of the United States National
-Museum, is apparently even more completely petrified than the
human bones from Vero. In Florida, mineralization of bones
or their inclusion in geological deposits has little chronological
significance.
The ‘‘fresh-water marl” that covers the deposits in the locality
of skeleton No. I is not found over the muck layer, or layer No. 3,
from which came skeleton No. II, but the point is immaterial. The
layer, except where exposed, is not or is but partly consolidated; and
even if it were solid it would have little bearing on the antiquity
of whatever may lie underneath. The writer found a very good
demonstration of this after he left Vero, on the Demere Key, off
Fort Myers on the west coast of Florida, and not very far south of
the latitude of Vero. He found there a low sand burial mound
the entire surface of which, consisting of sand, organic matter and
shells, materials gathered from the vicinity of the mound and from
the seashore, was consolidated to the depth of from four to sixteen
inches to such a degree that in places it was almost impossible to
penetrate it with a mattock. This “rock’’ included numerous
human bones, even skulls, a series of which is now in the National
Museum. Its age is possibly post-Columbian, for there were found
on the Key fragments of Spanish pottery and glass, while burial
sand mounds on neighboring keys yielded glass beads.
t Bureau American Ethnology Bulls. 33 and 52.
50 ALES HRDLICKA
In considering these problems the anthropological characteris-
tics of the bones themselves deserve serious consideration. They
now lie before the writer, and he has not found as yet a single
feature in which they would not agree with recent, more especially
Indian, bones. The juvenile or young adult incisor tooth presents
in a typical way the highly specialized characteristic form of the
Indian middle upper incisor; what there is of the lower jaw is
wholly of modern form; the skull of skeleton No. II by its lack of
thickness, good size, and subdued supraorbital ridges is actually of a
type superior to that of a large majority of the Florida Indians;
and the shape and dimensions of the other bones are those of a man
of the present day. There is nothing which would remind the
anthropologist of early man.
In conclusion the writer wishes to submit that besides all the
foregoing considerations there are broader anthropological and
archaeological problems which should receive due attention in all
cases of this nature. ‘They are both cultural and anthropological,
and their discussion must be reserved for the detailed report.
It may, however, be here briefly pointed out that an advanced state
of culture such as that shown by the pottery, bone implements, and
worked stone (brought from a considerable distance) implies a
numerous population, spread over large areas, acquainted thor-
oughly with fire, with cooking food, and with all the usual primitive
arts. Such a population would surely have left many tangible
traces of their presence on the Continent, some of which at least —
would by this time have been discovered.
It is the opinion of the writer, as the result of his investigations,
that the human bones found at Vero may well be prehistoric, and
date from the early part of the occupation of the Florida peninsula
by the Indians; but that no proof is furnished by the circumstances
of the find, or by the human bones themselves, which would
relegate the latter to an antiquity comparable with that of the
fossil remains with which they are associated.
ADDENDUM
While at Vero the writer obtained from Mr. Weills 20 frag-
ments of pottery recovered from the Vero deposits. In addition
to this, two fragments were obtained from the sand mound on the
FOSSIL HUMAN REMAINS AT VERO, FLORIDA 51
Indian River. This pottery was submitted for examination to
Professor Holmes, and his report follows:
December 1, 1916.
Dear Docror HRDiicKA:
I have examined with great care the pottery fragments obtained from the
site of the discovery of human remains associated with Pleistocene deposits
near Vero, Florida. They represent moderately small, undecorated vessels,
apparently simple bowls such as were in common use among the Indian
tribes of Florida. Compared with corresponding plain vessel fragments from
Florida sand mounds and from occupied sites generally, no significant dis-
tinctions can be made; in material, thickness of walls, finish of rim, surface
finish, color, state of preservation, and size and shape of vessels represented,
all are identical. There thus appears not the least ground in the evidence of
the specimens themselves for the assumption that the Vero pottery pertains
to any other people than the mound-building Indian tribes of Florida or to
any other than Columbian and immediately pre-Columbian time.
Sincerely yours,
W. H. Hotmes
Head Curator, Depariment of Anthropology
THE QUATERNARY DEPOSITS AT VERO, FLORIDA,
AND THE VERTEBRATE REMAINS CONTAINED
THEREIN
OLIVER P. HAY
Research Associate, Carnegie Institution of Washington.
I arrived at Vero on the evening of October 25 and left there
on October 31. Having examined with some care the geological
situation and having studied somewhat the vertebrate fossils
found in the strata designated by Dr. Sellards as No. 2 and No. 3,
I reach the following conclusions:
1. Stratum No. 2 was in general laid down during the Pleistocene.
—It seems hardly necessary to present arguments to sustain this
conclusion, for it is hardly probable that anyone will call it in ques-
tion. It is possible that some parts of the stratum were afterward
re-worked by the streamlet which flowed over it, but this was
accomplished during Pleistocene times.
2. The vertebrate fauna of No. 2 belongs to the Pleistocene, and
most of it is there by primary inclusion.—No place was discovered
from which the included bones and teeth might have been washed
in, nor do they in general have the appearance of transported
fossils. These bony remains are in what may be regarded as a
normal condition; as when, in a little valley furnishing food and
drink and shade, herbivorous and carnivorous species had resorted
and perished there for thousands of years. In a normal way their
bones have almost all fallen into dust. Some, buried under some-
what favorable conditions, endured longer, but softened and were
trampled into fragments by succeeding generations of elephants,
mastodons, horses, bisons, huge ground sloths, and smaller forms.
Only the most favored and protected bones and teeth have endured
to the present, mostly scattered, but sometimes remaining asso-
ciated with others of the same skeleton.
3. Al least the lower part of No. 3 is also of Pleistocene age.—
This deposit is somewhat more difficult to work for fossils, but it
52
FOSSIL HUMAN REMAINS AT VERO, FLORIDA 53
has furnished almost all the forms that are found in No. 2. It is
not improbable that some bones and teeth were redeposited from
the lower stratum, but not, I think, any considerable or essential
portion of them.
a) Considering the relatively small amount of erosion which
No. 2 suffered from the stream which laid down the muck bed,
there are too many fossils in the latter to permit the conclusion
that any great number of them came from the older deposit. The
lowest layers of muck early formed a blanket which protected the
sands of No. 2 from further disturbance.
b) The state of preservation of the fossils of No. 3 does not
indicate that they were redeposited from No. 2. They are not
more broken and waterworn than those of No. 2.
c) There are some extinct species in No. 3 whose remains
must lie where originally buried. The box-tortoise Terrapene
innoxia is found in both strata. Although the bones of the cara-
pace are usually co-ossified into one mass, this shell is so thin and
brittle that it would certainly have fallen into pieces on being
rolled along a stream bed. It is even now extremely difficult to
unearth a shell without breaking it. Yet one whole carapace and
large portions of others have been secured from No. 3. From this
muck bed there come seven bones of one individual of an extinct
snapping tortoise, probably Chelydra sculpta. The shell of this
animal, like that of our living species, is thin and loosely articulated.
On maceration the bones separate easily. Had the seven bones
referred to been buried originally in No. 2, they would, on being
washed out, have been scattered like autumn leaves.
d) In No. 3 there is a deer of the genus Odocoileus which is
smaller than the one found in No. 2.7. From No. 3 Dr. Sellards
has sent me a. fifth cervical vertebra which shows that this deer is
very distinct from the existing Virginia deer and still farther
removed from the mule deer. The fox referred with doubt by
Dr. Sellards to the red fox is certainly an undescribed species,
having had a heavier lower jaw than that of the red fox. A
femur from No. 3 probably belongs to the same species. It is
larger, straighter, and more flattened than that of the red fox.
tSellards, Sth Ann. Rep. Fla. Geol. Surv. p. 1409.
54 OLIVER P. HAY
In short, there are so many well-preserved extinct vertebrates
in No. 3 that it must be referred to the Pleistocene; and the study
of the collections adds continually to the number.
4. A few words only about the human bones.—I consider now only
those found at the locality illustrated by Sellards’ Text-Fig. 6 and his
Plate 16 and Plate 17, Fig. 2. Had no human bones been found
there the following explanation would, I think, hardly be questioned.
No. 2, consisting mostly of sand, had been deposited, leaving traces
of horizontal stratification. At a later time the swollen streamlet
cut down through it to the underlying marl. About four feet
away at the same time it cut down nearly to the marl. The two
currents left a ridge of undisturbed sand which contained some
bones. As the currents lost their force, sand began to be deposited
on the sides and summit of the ridge. Had there been any con-
siderable interval, this ridge of sand would have been flattened
down and disturbed in various ways. Before the freshet spent
itself a mass of vegetation was swept down and deposited, mostly
in the channels but partly on the ridge, thus sealing it in until our
day. As to the human bones found lying on the slope of No. 2, a
reasonable explanation is that they had previously been scattered
and inclosed in its sands and then laid bare by the freshet. Their
condition of fossilization is the same as that of the animal bones
found near by, and their broken condition indicates that they had
suffered from the trampling of animals, as those other bones had.
5. The age of stratum No. 2 and of at least the lower pari of No. 3
is not later than middle Pleistocene-—The fauna afforded by the
deposits In question is essentially that which is found in the Aftonian
interglacial beds in Iowa and in the Equus beds of the Plains. From
the latter it may be followed into Texas, thence eastward into
Florida and South Carolina. Of this fauna two species of ele-
phants, the common mastodon, Megalonyx, and the giant beaver,
continued on until after the Wisconsin glacial stage. Other species,
the saber-tooth tigers, Equus complicatus, the tapirs, most of the
extinct bisons, and Mylodon probably disappeared before the Wis-
consin. In the earlier Pleistocene deposits only are found Elephas
imperator, camels, several species of horses, and many edentates.
At Vero have been found three species of horses, at least four
FOSSIL HUMAN REMAINS AT VERO, FLORIDA 55
edentates (including Mylodon), and a camel. Chlamytherium
was originally found on Peace Creek in deposits which were then
supposed to be Pliocene. In the same deposits was found a jaw
containing a tooth of an elephant which is quite likely E. imperator.
This species has not yet been found in No. 2 at Vero, but about
three miles west of the place Sellards found a lower jaw which be-
longs probably to this species. It is known from Dallas County,
Alabama, and from Charleston, South Carolina. The writer regards
- it and camel remains as particularly indicative of the Aftonian fauna.
It is possible that this fauna continued on for another stage
or two without great change, but by the time of the Illinoian drift
it had become essentially modified.
6. The human bones appear to be of Pleistocene age.—At present
I perceive no other reason for doubting this than that their presence
in No. 2 and No. 3 contravenes our present ideas regarding the
history of the human race.
ARCHAEOLOGICAL EVIDENCES OF MAN’S ANTIQUITY
AT VERO, FLORIDA
GEORGE GRANT MacCURDY
Yale University
The apparent association of human remains and artifacts with
fossil animal remains in Pleistocene deposits is always and every-
where sufficient to challenge the attention of scientists. This
is especially true of the New World, where Pleistocene man has
not yet won a place in the prehistoric hall of fame; hence the wide
interest taken in the announcement by Dr. E. H. Sellards," state
geologist of Florida, that he and his colleagues had found such an
association at Vero, Florida.
As one of several invited to investigate the circumstances of
the find on the spot, the writer obtained leave of absence from
Yale University for this purpose, and visited the Vero site during
the week of October 23~29 as the guest of Dr. Sellards. To him
and to his assistant, Mr. H. Gunter, as well as to his local associates,
Messrs. Frank Ayers and Isaac M. Weills, grateful acknowledg-
ments are due for facilities so generously extended. ‘The writer’s
visit approximately coincided with those of Dr. Rollin T. Chamber-
lin of Chicago and Drs. O. P. Hay, A. Hrdlicka, and T. Wayland
Vaughan of Washington, D.C. The headquarters of the party were
at the site itself, one-half mile north of the village of Vero, and
easily reached by the highway that parallels the railroad tracks.
The drainage canal which cuts through the site is of itself
sufficient proof of the flatness of the country. The human remains
and artifacts and the fossil animal remains were all found at. the
junction of two lateral valleys, which united to form the trunk of a
wider valley; in this valley until recently a stream followed an
“7]|-defined, anastomosing, and frequently changing channel.” At
t Amer. Jour. Sci., XLII (July, 1916); Eighth Ann. Rep. Fla. State Geol. Survey,
pp. 121-60, 1916; Science, N.S., XLIV (October 27, 1916).
56
FOSSIL HUMAN REMAINS AT VERO, FLORIDA 57
this junction the canal enters from the west the main-stream
valley, which it follows for some 800 feet.
Along both banks of the canal Dr. Sellards had prepared sections
for the inspection of the party. One of these sections was extended
by Dr. Hrdlicka, who also opened up a new section along the east
bank of the tributary canal that follows the course of the lateral
valley entering from the south. Additional animal remains were
found daily during the stay of the party, especially in the middle
one of the three strata described by Dr. Sellards. As to the cor-
rectness of his interpretation of the stratigraphic section there
would seem to be little doubt. It remains to be seen whether all
his conclusions can stand the test with equal success.
Dr. Sellards had brought with him from Tallahassee human
remains found to date in stratum No. 2 and along the contact line
between it and stratum No. 3, also certain flint chips, bone imple-
ments, the tip of a proboscidian tusk, and a fragment of a bird bone
—the last two with markings which he believed to have been
made by tools. These were all carefully studied by the writer
while he was at Vero. Later the human bones were sent to
Dr. Hrdlicka at the National Museum and will be the subject of
his contribution. From a study of them at Vero before the broken
parts were assembled, and without material at hand for comparison,
the writer agrees with Dr. Hrdlicka that they are in no way different
from Indian skeletal remains found in the sand mounds of Florida.
In the writer’s opinion the markings on the tip of the proboscidian
tusk and on the fragment of bird bone, both from stratum No. 2,
are not the work of man.
A consignment including flint chips and implements, bone
- implements, and an ornament and potsherds were sent to New
Haven after the writer’s return. The sherds and some of the
other objects are from stratum No. 3. Some of these specimens
were figured by Dr. Sellards; certain of the figures which seemed
to be inadequate in Sellards’ work are reproduced herewith.
The flint spall, No. 6964 (Sellards’ Text-Fig. rr), was found in
stratum No. 2, in the south bank, 460 feet west of the railroad bridge
and 3 feet from certain bones of human skeleton No. II (Fig. .z).
Another and smaller spall of identical material, which might well
58 GEORGE GRANT MacCURDY
have been chipped from the same parent block, was, according to
Sellards, found in the south bank 460 feet west of the bridge, but in
stratum No. 3 (Fig. 2). That of these two chips of like material
and so near each other in respect to horizontal displacement one
should have been found in stratum No. 3 and the other in stratum
No. 2 is significant. The question arises whether both might not
have been originally in stratum No. 3, one having worked its way
down into No. 2 by the aid of growing roots or burrowing animals.
While Dr. Sellards does not recall having seen any roots reaching
into stratum No. 2 where the spall reproduced in Fig. 1 was found, he
admits that roots do penetrate this stratum in places, notably a
Fic. 1 Fic. 2 Fic. 3
Fics. 1-3.—(1) Flint spall from stratum No. 2, south bank, 460 feet west of the
bridge and near human bones; (2) flint spall of identical material from stratum No. 3,
south bank, 460 feet west of the bridge, from siftings; (3) flint spall from stratum No. 2,
south bank, 460 feet west of the bridge, from siftings (+). Nos. 6964, 7072, and 7049.
little farther west where flint No. 7055 (not herein figured) was
found.
These spalls were never retouched or utilized. Each has what
the French call a plan de frappe (“plane of percussion”) and a
well-marked bulb of percussion. ‘The inner or conchoidal surface
is fresh and the edges are unworn. ‘They were evidently chipped
from the parent block not far from where they were found. At one
time the presence of a bulb of percussion was looked upon as a sure
sign of human agency. Certain rare examples from the base of
the Eocene at Belle-Assise, Clermont (Oise), and from the Oligocene
at Boncelles, Belgium, are proof that the bulb is not an infallible
sign. By accidentally letting one flint fall upon another, the writer
has on one occasion unintentionally caused the production of a bulb
of percussion. It is, however, quite logical to assume that the vast
FOSSIL HUMAN REMAINS AT VERO, FLORIDA 59
majority of chips with bulbs that occur in Pleistocene and later
deposits have been produced intentionally, especially when asso-
ciated with human skeletal remains or with undoubted artifacts.
This is doubly true at Vero, because the source of the flint is the
Ocala or the Tampa formation a hundred miles to the northwest of
Vero. The cores from which the chips were struck could not well
have been transported that distance over so flat a country except
through human agency.
The small flint chip reproduced in Fig. 3, and thought by
Sellards (his Text-Figs. 7 and 8) to be an implement, is likewise only
a chip or spall with its plane of percussion and bulb of percussion.
The multiple facets on its back or outer surface are due to the fact
that it was an inner instead of a superficial chip. It also is from
the south bank 460 feet west of the bridge, hence from near skeleton
No. II and the other two spalls here reproduced. While obtained
from siftings, it is believed by Dr. Sellards to have come from
stratum No. 2. In a recent letter he emphasizes the fact that
“up to the present the number of spalls taken from stratum No. 2
is in excess of the number taken from stratum No. 3, notwithstand-
ing that rather more material from No. 3 has been handled, nd
fully as much material from that stratum has been passed through
the sieve as from stratum No. 2.” This fact, however, would
not seem to have any very direct bearing on the question whether
or not flints from stratum No. 3 had worked their way down into
stratum No. 2.
A typical arrowhead of flint with barbs and stem, the latter
however broken off, came from the contact line between strata
No. 2 and No. 3 in the south bank 470 feet west of the bridge
(Sellards’ Fig. 1, Pl. 21).
For the sake of comparison bone implements from strata No. 2:
and No. 3 are reproduced in Figs. 4-6. Fig. 4 is a typical point from
stratum No. 3, south bank, one of several from 450 to 470 feet west
of the bridge. The fragment of a similar point, obtained in siftings
from stratum No. 2, south bank, 462 feet west of the bridge, is
shown in Fig. 5. Another and nearly complete point, obtained
in siftings from stratum No. 2, south bank, 480 feet west of the
bridge, differs from the other two only in size (Fig. 6).
60 GEORGE GRANT MacCURDY
So far as the writer is aware no potsherds have as yet been
reported from stratum No. 2, although they occur somewhat plenti-
fully in stratum No. 3. Of the dozen sherds sent to New Haven
every one is more or less waterworn. When subjected to stream
action, these sherds would show the effects of wear quicker than
would the bones, flints, and bone implements. The pottery is of
fairly uniform quality, the paste being neither crude nor fine. Itis
black to brown in color and the walls are of medium thickness.
Judging from these twelve sherds, the ware was unpainted and un-
decorated. Of the three rim fragments, two are from bowls of
Fic. 4
Fic. 6
Fics. 4-6.—(4) Bone point from stratum No. 3, south bank, 450-70 feet west of
bridge; (5) fragment of bone point from siftings of stratum No. 2, south bank, 462 feet
west of the bridge; (6) bone point from siftings of stratum No. 2, south bank, 480 feet
west of the bridge (}). Nos. 6912, 6963, 6981.
medium size, the third, somewhat thicker, is from a medium-sized
bowl with slightly incurved rim. All these rims are plain but
carefully finished. The smoke stains and accumulated soot indi-
cate that these were culinary vessels. It should be recalled that
the sherds, flints, and bone implements of stratum No. 3 are found
in the north as well as the south bank of the canal at the junction of
the two lateral valleys previously mentioned. None of these differ
from similar antiquities found on the surface or in Florida mounds.
To summarize the archaeological evidences of man’s antiquity
at Vero, one can say that the pottery, bone implements, including
fishhooks, bone heads, and flint arrowheads from stratum No. 3
and from the surface of contact between it and the stratum below,
FOSSIL HUMAN REMAINS AT VERO, FLORIDA 61
all point to a period that might well have continued down to the
close of the prehistoric period in Florida. This is also true of the
human skeletal remains from the third stratum. On the other
hand, of the 25 mammalian species from the second stratum as
listed by Dr. Sellards, ten, including Elephas columbi, Mammut
americanum, Equus leidyi?, and Tapirus haysit?, recur in stratum
No. 3. Assuming that the stratigraphy is not misleading, the
conclusion is either that this particular phase of the Neolithic
period in America dates back farther than many had supposed,
or else that certain fossil mammals continued to live on in Florida
until a comparatively recent date.
The chief interest centers in the second stratum. From it no
undoubted stone implements have thus far been reported. Al-
though probably produced through human agency, the flint spalls
from this deposit do not differ from those in the deposit above, in
one case there being absolute identity of material. While a greater
number of bone objects have been found in the third deposit than
in the second, bone points of the same type occur in both; neither
do these seem to differ as to their chemical state. Potsherds,
fairly frequent in stratum No. 3, have not yet been reported from
the stratum below. Of the human skeletal remains there does not
seem to be any appreciable differentiation between those from the
second and those from the third stratum.
There are to be noted then the absence of well-defined stone
artifacts and of pottery from the second deposit; the presence of
both in the third; the similarity of the flint chips from the two
deposits; the similarity of the bone points in both deposits; and the
greater number and variety of bone artifacts including ornaments
in the third deposit. But for the similarity of the flint chips and
the bone points the cultural evidence is very much as one might have
been led to expect, assuming of course that the stratigraphy is
unmixed and that all specimens have been found zm sztu. On the
other hand, in the absence of stratigraphy as a guide, of all the
human and cultural remains reported from stratum No. 2 none
would seem out of place in stratum No. 3.
It will be recalled that one flint spall (see Fig. 3) referred to the
second stratum was from siftings; and that the two bone points
62 GEORGE GRANT MacCURDY
(see Figs. 5 and 6) referred to the same stratum are likewise from
siftings. Even if these were eliminated, there would still remain
as stratigraphically troublesome elements the two flint chips (see
Figs. 1 and 2). The presence of plant stems, acorn cups, and pieces
of wood in the second stratum, although by no means so abundant
as in the third stratum, nevertheless give to it an aspect of com-
parative newness. Some of the leaves in the muck at the base of
the third stratum look as if they might have been buried only a
few years ago.
From observations made on the spot and from a study of speci-
mens submitted, the writer is of the opinion that for the most part
the human skeletal remains, flint chips, and artifacts probably found
their way to this meeting-place of waters through the same agencies
as did the various animal and plant remains, and that there has been
more or less dovetailing of the two deposits, because of the peculiar
location of the site at the junction of two streams coming from oppo-
site directions. If these premises be true, it would be hazardous
to attribute any great antiquity to even the oldest human and
cultural remains from Vero. It would be more logical to assume
that some of the extinct forms found in the second stratum are
perhaps derived from an older deposit; that others lived on in
that southern clime longer than has hitherto been supposed, and
that the presence of the Indian hunter had much to do with the
final ringing down of the curtain on the drama of their ultimate
extinction.
SUGGESTIONS FOR A QUANTITATIVE MINERALOGICAL
CLASSIFICATION OF IGNEOUS ROCKS
ALBERT JOHANNSEN
University of Chicago
It is with considerable hesitation that the writer introduces a
new classification of igneous rocks. He knows that he who adds
a single term to an already overburdened vocabulary is looked upon
with disfavor, while he who brings in many has hearty objurgations
heaped upon him; yet he hopes, as others who have gone this way
before him have hoped, by fixing definite boundary lines beyond
which the different families cannot pass, to eliminate the multiplica-
tion of names for rocks which differ in no essential particulars from
previously described types.
It is being recognized, more and more, that there is need for
three classifications of igneous rocks. Of these, one must be for
field use’ and megascopic. Another must be chemical, after the
manner of the systems of Osann? and C.I.P.W.3 The third must
be mineralogical. The old classifications of Rosenbusch and Zirkel
are more or less mineralogical, it is true, and are not to be discarded
lightly, but they fail especially in their lack of the quantitative
element. Furthermore, they are neither purely mineralogical,
purely chemical, nor purely geological. For example, certain
dike-rocks are classified by Rosenbusch, on the basis of their field
associations with nephelite-syenites, essexites, etc., as rocks of the
alkali series, and to them he gives specific names, yet they are miner-
alogically and chemically identical with normal rocks of the alkali-
lime series. He depends in part, therefore, on field associations
«Field classifications are given by Cross, Iddings, Pirsson, and Washington,
Quantitative Classification of Igneous Rocks (Chicago, 1903), p. 180; L. V. Pirsson,
Rocks and Rock Minerals (New York, 1908), p. 202; Albert Johannsen, “ Petrographic
Terms for Field Use,” Jour. Geol., XIX (1911), 317-22. A revised form of the latter
will appear shortly.
2A, Osann, Tschermak’s Mitteilungen, XIX, XX, XXI, XXII (1899-1903).
3 Cross, Iddings, Pirsson, and Washington, of. cit.
63
64 ALBERT JOHANNSEN
for classification. Elsewhere he uses chemical data to classify
rocks which he defines in mineralogical terms; for example, the SiO,
percentage must lie between certain limits, or the sum of the alkalies
must be less than the alumina, etc.
Another objection to the present system of classification is the
fact that rock terms have been used loosely or with different
@ meanings. ‘Thus dolerite,
originally applied to a
coarse basalt, has been used
for any dark rock, and in
England is used for rocks
which we call diabases.
The term diabase in the
United States means a
dike-rock with an ophitic
texture, yet it was origi-
nally used for Paleozoic
basalts and is still so used
1 various countries:
Fic. 1—One hundred and nine so-called Basalt has been applied
‘““sranites.” Open circles are rocks of Class 1, to plagioclase rocks with
and dark circles rocks of Class 2. The double augite and olivine irr espec-
circle is the mean of Daly’s granites re- : ;
computed into the probable modal minerals. tive of the kind of feldspar )
to labradorite-pyribole™
{) \
LX/\
[XS IAAI W NAVAVaNN
ake /N\IiN/\¥,
rocks with or without olivine, to the darker labradorite-pyribole
rocks, to post-Tertiary extrusives of gabbroic magma, etc.
The loose usage of terms by different writers with respect to
the mineralogical compositions of rocks is well brought out by
Figs. 1 and 2, in which the three corners of the triangles represent
respectively quartz, potash-feldspar, and plagioclase. In Fig. 1
are plotted tog so-called “granites,” taken, not from old descrip-
tions, but from comparatively recent ones in which the actual
mineral compositions were determined by the Rosiwal method
or by careful estimation by the various writers themselves. In a
few cases the rocks were doubtless named on the basis of their
t A general term for the members of the pyroxene and amphibole groups (Albert
Johannsen, of. cit.).
MINERALOGICAL CLASSIFICATION OF IGNEOUS ROCKS 65
chemical compositions, but in most cases their chemical composi-
tions are as far from true granites as are their mineral compositions.
The figure shows that there are actually 6 potash-granites (one of
them quartz-rich), 63 normal granites (4 of them quartz-rich),
29 quartz-monzonites (1 of them quartz-rich), and 11 granodiorites.
Fig. 2 represents 30 so-called ‘‘syenites.’”’ There are 2 potash-
syenites, 3 normal sye-
nites, 14 normal granites,
3 monzonites, 7 quartz-
monzonites, and 1 grano-
diorite.
Many recent papers
show the tendency toward
a quantitative mineralogi-
cal classification. Thus
Brogger proposed
fairly definite boun-
daries for monzonite
and quartz-mMonzo- ré
nite. From the latter Fic. 2.—Thirty so-called ‘“‘syenites”’
Lindgren separated _
granodiorite, and established limits so clearly that almost all rocks
described as granodiorites are actually such. But covering a wider
field are later papers by Iddings' and Lincoln.? Each of these writers
proposed a definite classification, and more recently Shand’ sug-
gested subdividing rocks according to their percentages of light
and dark constituents. To the writer, none of these classifica-
tions appears so satisfactory as that which he has presented to
his students, with various modifications, during the past seven
years. The system was first thought out in the summer of 19009,
and even so long ago as the summer of to1o the writer pre-
pared plaster models of tetrahedrons, cut into subdivisions essen-
tially as shown here. Owing to press of other work and lack of
t Joseph P. Iddings, Igneous Rocks (New York, 1913), Vol. II.
?Francis Church Lincoln, ‘‘The Quantitative Mineralogical Classification of
Gradational Rocks,” Econ. Geol., VIII (1913), 551-64.
3. J. Shand, “‘A Recording Micrometer for Geometrical Rock Analysis,” Jour.
Geol., XXIV (1916), 404.
66 ALBERT JOHANNSEN
sufficient data in the literature as to the modes of rocks, the publica-
tion was delayed. In the present paper the writer presents the
system in a tentative form, hoping to receive from other petrog-
raphers expressions of opinion and suggestions for modifications.
Later he hopes to show the relationships, both mineralogical and
chemical, existing between the rocks falling into the various groups.
The system here proposed is strictly mineralogical, quantitative,
and modal, and is directly applicable to all plutonites and to prac-
tically all extrusives. The writer’s objections to the percentage
values set by various other authors will be given below. Appar-
ently the dividing lines have previously been arbitrarily selected,
and no attempts have been made to gather published data with
respect to the modes of rocks. ‘There is, in fact, a surprising lack
of such data, the writer having been able to find published reports
of less than 600 quantitatively determined rocks.
If the reader has ever attempted to find, from the average
report, the relationship existing between a newly described rock and
the older types, he will in many cases have found it impossible.
This is clearly shown by the fact that Rosenbusch himself, by the
misinterpretation of descriptions, has misplaced rocks, grouping
them with totally unrelated types. If the reader will turn at
random to almost any petrographic report,’ and will read a descrip-
tion and then attempt to picture to himself the rock described,
in most cases he will find that owing to lack of quantitative data
no idea as to its appearance can be obtained.
A name should convey an idea as to the character and appear-
ance of a rock, and it should not be necessary, as it now unfortu-
nately is, for one to read the description of a rock to know what a
writer means. So far as the name itself is concerned, it is of slight
importance, provided the texture is described and accurate quantita-
tive details of the average rock are given. But without quantitative
details serious errors may arise. Thus in a recent petrographic
report a rock was said to contain orthoclase, andesine, quartz,
«The writer is guilty of having written, indefinite descriptions himself. As an
exception to the general rule of poorly written and indefinite reports, he likes to refer
his students to Dr. H. S. Washington’s ‘““Roman Comagmatic Region,’ Carnegie
Publication No. 57. Here there is never the least doubt as to the mineralogical
composition and appearance of a rock.
MINERALOGICAL CLASSIFICATION OF IGNEOUS ROCKS 67
biotite, and hornblende, and was called a syenite. One naturally
would suppose from the name that andesine and quartz were of
subordinate importance, yet an examination of many thin sections
showed 20 per cent quartz and 30 per cent each of orthoclase and
andesine, a rock which is a quartz-monzonite BROGGER. One rock
found to be thus incorrectly named raises doubts as to the accuracy
of the determinations of all other rocks in the same report.
Wg 3
Fig. 4 Fig,5
Fics. 3-8.—Various proportions of dark minerals in a rock
During the past few years the writer has required his students,
in their rock descriptions, to give the percentages of the different
constituents,* and he has invariably found that the estimates
of the less abundant minerals, such as the dark constituents in leuco-
cratic rocks or the light constituents in those that are melanocratic,
are entirely too high, and that the‘first summation of all the con-
stituents runs between 80 and g5 per cent. The reader may test
for himself, before reading farther, his ability to estimate per-
centages by examining Figs. 3 to 8, which were made by pasting
‘For a specimen card showing percentages see Albert Johannsen, A Manual of
Petrographic Methods (New York, 1914), p. 614.
68 ALBERT JOHANNSEN
into circles of known size irregular fragments cut from pieces of
black paper which bore definite ratios to the circles. Of course, if
one has often measured constituents by the Rosiwal method his
estimates are likely to be fairly good.
The system here presented is not intended as a substitute for any
chemical system. But, as so well expressed by Clarke, ‘‘ Even if it
[the C.I.P.W. system] should be finally adopted by all petrologists,
some form of classification like that now in vogue would have to be
retained with it. Good analyses cannot be obtained for every
rock which the geologist is called upon to determine, and in many
cases he must be content with the results of a microscopic examina-
tion.”? And it-is also true that for rocks which show considerable
decomposition the microscopic method is far more likely to give
good results than the chemical.
As an objection to a quantitative mineralogical system, such as
is here proposed, it will be said that it is not always possible to
determine the exact composition of rocks with a glassy base or
extrusive rocks of the alkali series. But the percentage of inde-
terminable rocks is comparatively small, and for these there still
remain, if necessary, chemical methods for determining the compo-
sition of the base. Most glassy rocks are leucocratic, and a recalcu-
lation into the minerals which would have crystallized had the
conditions been right is easy. Since the majority of these glassy
rocks are rhyolitic, one is no worse off in adopting a quantitative
classification than at the present time, when they are called rhyo-
lites from microscopic examination. In such cases it would not
be objectionable to make use of tentative names which could be re-
vised after chemical analyses have been made. Ina later paper the
author hopes to present a method for determining quantitatively
even these rocks with very little difficulty. Certainly 95 per cent
of fresh igneous rocks can be classified microscopically. When
rocks are completely decomposed, no determinative system,
chemical or mineralogical, will help.
The plutonic rocks must necessarily form the type families of
any mineralogical classification of igneous rocks, and extrusive
* The actual percentages in the figures are 3, I, 2, 5, 10, and 50.
2¥. W. Clarke, ‘Data of Geochemistry,” U. S. Geol. Surv., Bull. 616, (Wash-
ington, 1916), Pp. 432.
MINERALOGICAL CLASSIFICATION OF IGNEOUS ROCKS 69
and hypabyssal rocks must be regarded as modifications of these.
In this paper the writer has given names only to the plutonic
representatives of the few families considered, it being understood,
of course, that the granite family includes rhyolites; the syenite
family, trachyte; the monzonite, latite; etc.
The basis ofethe classification here proposed is a double tetra-
hedron (Fig. 9), each trihedral angle of which represents certain
mineral constituents. If there
were a geometrical figure hav-
ing ten or twelve corners, each
equally distant from each of the
others, it would have been pos-
sible to use a single mineral at
acorner. Since there is no such
figure, and rocks must be lo-
cated with reference to all of
the minerals which occur in
them, it was found necessary
to divide the minerals into as
many groups as there are cor-
ners in a tetrahedron. But
quartz and the feldspathoids never occur together, so it was possible
to make the classification in five dimensions by using two tetrahe-
drons with a common base (Fig. 9). This arrangement was found
to answer the purpose admirably, for the relationships between
rocks which may contain either quartz or nephelite, etc., and which
appear anomalous in the old classifications, are clearly shown.
The groups of minerals represented by the corners of the double
tetrahedron are: (1) quartz (symbol Qu’); (2) potash feldspars
(symbol Kf), including the orthoclase molecule in anorthoclase;
Fic. g.—Subdivisions of the double
tetrahedron into classes.
tIn the figures in this paper the quartz corner is indicated by the symbol Qu.
The letter ‘“f” is used for feldspar, therefore Kf indicates the potash-feldspars—
orthoclase, microcline, and the orthoclase molecule in anorthoclase; Naf indicates
albite and the soda molecule in anorthoclase, while CaNaf represents the acid plagio-
clases and NaCaf the basic plagioclases, the element in excess being given in italics
temporarily to avoid confusion, although there need be none if one thinks of the symbol
as reading calcium-bearing soda feldspar for the acid plagioclase and soda-bearing
calcium plagioclase for the basic. Caf is used for anorthite, and Foids for the feld-
spathoids, lenads being unavailable from its use for certain normative minerals of the
C.I.P.W. system.
70 ALBERT JOHANNSEN
(3) all plagioclases and the albite molecule in anorthoclase; (4) all
feldspathoids; (5) the mafites,! including the ferromagnesian con-
stituents, the ‘‘ores,”’ etc., as given below.
As shown in Fig. 9, the double tetrahedron is unsymmetrically
divided on certain faces by the traces of planes parallel to the
quarfeloid? faces; on others, by lines parallel to one side as well as
by lines converging to one of the angles. Experiments were made
with symmetrical divisions of various kinds, but it was found
impossible to fit the rocks as now named into compartments so
made. It is true that new names might have been devised for such
subdivisions, but it was not thought desirable to discard entirely
the old and well-tried classifications which have very much to recom-
mend them besides the fact that they have been so longinuse. The
old classifications are unsymmetrical, for we speak of a rock as a
quartz-syenite, quartz-monzonite, quartz-diorite, etc., when it
contains any amount of quartz. With respect to this mineral, there-
fore, the classification is based upon its ratio to the sum of all the.
other constituents, and the lines of division must be parallel to a
side of the tetrahedron. The same is true also of the feldspathoids.
In the divisions according to the feldspars, however, we find, for
example, that a rock is a quartz-monzonite whether the percentage
of feldspar among the light constituents is to or go. Here the
divisions are based upon the ratio of the feldspars to each other,
irrespective of what their amount may bein the rock. The division
lines, therefore, must converge toward the quartz and feldspathoid
corners, as shown in Fig. 9.
t When the writer proposed (Jour. Geol., XTX [1911], 319) the term ‘‘femag” asa
substitute for ferromagnesian minerals which are not minerals of the norm, he did not
stop to consider its euphony or whether it fitted into the C.I.P.W. terminology, but
thought of it only as a term to take the place of ‘‘femic,’’ which was being misused.
He is perfectly willing to substitute “‘mafic” as an adjective, as proposed by the authors
of the C.I.P.W. system (Jour. Geol., XX [1912], 561). He wishes to use here a term for
all the dark minerals of a rock except those that are pneumatolytic, and therefore uses
‘“‘mafite’’ as a noun, feeling at liberty to include in it, since the word has not been
used before, certain iron minerals, as listed below.
2 C.1.P.W. suggest “‘felsic’’ as an adjective for the minerals quartz, feldspars, and
feldspathoids. The writer here uses ‘‘quarfeloids’” (QUARtz, FELdspar, feld-
spathOIDS) as a noun for these minerals in the front faces of the double tetrahedron,
“felsite””’ being unavailable from its use as a rock name. ‘‘Leucocrates’’ cannot be
used, since all light-colored minerals are not included.
MINERALOGICAL CLASSIFICATION OF IGNEOUS ROCKS 71
The igneous rocks may be divided into various classes according
to the percentage of dark constituents present. Any number of
divisions might, of course, be made; Shand" proposed twelve,
though more for descriptive purposes than classificatory. It is,
however, not desirable in a classification to multiply excessively the
number of classes into which the rocks are divided, and they may
be gathered into rather large groups. ‘Tentatively four classes
have been made: (1) rocks with less than 5 per cent of dark
constituents, (2) dark constituents
between 5 and 50 per cent, (3) dark
constituents between 50 and g5 per
cent, and (4) dark constituents
more than 95 per cent. Now since
these division lines represent planes
parallel to the two quarfeloid planes
(quartz-feldspars and feldspars-
feldspathoids), Fig. 9, they form
similar triangles whose sizes repre-
sent the amounts of light con-
stituents, decreasing with increase Pen ag le ene le ad uk
in dark constituents and approach _ ondary double tetrahedron into orders.
to the mafite corner. For conven-
ience, however, since they are similar they may be represented by
triangles of the same size.
Thus far the classification is one of five dimensions. But this
is not enough. The kind of plagioclase in the rock must be taken
into consideration. To bring this factor into the classification,
imagine the lozenge-shaped quarfeloid plane to consist of two
sheets of paper fastened together only along the Qu-Kf-Foids edge.
If now the loose corners of the two sheets be separated a distance
equal to a side of the original triangle, a new double tetrahedron
will be developed, the horizontal line along which it was opened
representing all plagioclases, the ends being formed by the Ab and
the An molecules (Fig. 10). The same thing can be done, of course,
with the double triangles representing the other classes, and the
classification will now be made up of four double tetrahedrons,
tS. J. Shand, op. cit., p. 404.
72 ALBERT JOHANNSEN
one for each class, the corners being formed by quartz, potash-
feldspar, albite, anorthite, and the feldspathoids. But these
tetrahedrons may be subdivided into orders. Based on the old
classifications, these orders depend upon the proportions of the albite
to the anorthite molecule; consequently the divisions must be made
by planes all of which cut the quartz—potash-feldspar—feldspathoid
edge but separate across the central plane of the double tetrahedron,
as shown by the dotted
lines in the figure, or by
Fig. 11, which is a hori-
zontal section through the
center. ‘That is, the edge
Qu-Kf-Foids remains
common to all of the di-
visions, the plagioclase
corner simply having been
changed. Now while the
triangles formed by the
intersections of
these planes with
Nar the tetrahedron
Kf
Fic. 11.—A section through the central plane of Fig. 10 (F ig Q 10) are not
all equilateral, the
relative position of any rock plotted on an equilateral triangle on the
basis of the three components represented by its corners and reduced
to 100 will be the same as the same rock plotted with four components
within the solid tetrahedron. Consequently the different orders also
may be represented simply by a series of double equilateral triangles
(Figs. 20-23 or 24-26) whose right-hand corners vary with the kind
of feldspar. It would, of course, be possible to make 20 or 100 or
more different orders based upon variations of 5 or 1 or some other
percentage in the albite content, but this is neither desirable nor
necessary. Here the divisions have been made (1) albite (Abyo.An,
to Ab,;An;), (2) oligoclase and andesine, (3) labradorite and
bytownite, (4) anorthite (Ab;An,; to Ab,Anioo), giving four orders.
In other words, the dividing points between albite and anorthite
are 100-95-50-5-0 of the albite molecule.
MINERALOGICAL CLASSIFICATION OF IGNEOUS ROCKS 73
There are now six dimensions in the classification, and since
each pigeonhole will represent not only a plutonic rock but also a
hypabyssal and an extrusive, we may say we have a classification
in seven dimensions, yet every rock may be shown by a single
point on a drawing ina single plane. The more detailed description
which follows may make this clearer.
NUMBER AND POSITIONS OF THE VARIOUS DIVISION LINES
Classes.—The dividing lines between the various classes, orders,
families, etc., were not selected at random, but an attempt was
made to see if they have any logical positions. For this purpose
the writer has been collecting data on cards for all rocks whose
modes in mineral percentages have been determined. ‘The number
is small, less than 600 such rocks having been found. Unfortu-
nately this number is too small to determine definitely all points,
but the writer found that in most cases preliminary graphs with
fewer analyses showed practically the same curves as the ones
here given.*
In order to determine the positions of the dividing planes
between the light and the dark rocks, and to decide whether there
should be four or five classes (namely white, light, medium, dark,
and black), the rocks of the various families were plotted in Fig. 12,
in which the abscissae represent the proportions of light constitu-
ents in the rock and the ordinates the number of rocks whose
modes were known, the percentages being gathered by fives to make
a smoother and more representative curve than the individual
percentages would have made. The lower curve in the figure is
the curve of all rocks (585) of which the writer had the modes, and
includes the alcalic rocks as well as the families given in the upper
curves. All the curves except the one for gabbro, which does not
extend so far, show an increase at go-95 per cent light and a
decrease beyond that toward 100 per cent. Consequently rocks
may well be called leucocratic when there are 95 per cent or more
of light minerals; and there is no objection to making the melano-
cratic division beyond 95 per cent dark. A difficulty appears in
«Since this paper was written, 91 additional mode-analyses have been found,
but the graphs remain practically as they were.
74 ; ALBERT JOHANNSEN ©
making a third division. In the granite and syenite, monzonite
and quartz-monzonite, syenodiorite and granodiorite families a
line separating 50 per cent light from 50 per cent dark would
throw practically all of the rocks on the same side. With respect
to diorite and quartz-diorite the curve is not good, owing to
insufficient data, and it shows no definite maximum. ‘The gabbro-
curve has its maximum at 60 per cent light. With the gabbros
5
Ga ldgasceeeesedeeece
ST SG ae
Se Nz
i is N
is
Ee
ae
athe
N
aa |
jes
a
seta
EERE
EMS
ie
=
ACRE
Z
iN
zi
WA
te
CHE
HERE
a (a aha Eaan ae 5
| et
(SSeEeees = SEEN
[istore [a [esas DZ 50
Fea oe ae taf)
ia sa |e 2
dark 0 5 10 20 30 40 50 60 70 8
Percentage of light minerals
fo)
ve}
°o
=)
fo}
oO
xf
be
g
ct
Fic. 12.—Curves showing the number of rocks with various percentages of
light and dark constituents: A, granite and syenite; B, quartz-monzonite and monzo-
nite; C, granodiorite and syenodiorite; D, quartz-diorite and diorite; EL, quartz-gabbro
and gabbro; F, all rocks.
and diorites it might be better to make five classes with dividing
lines at o—-5—35—-65—95-100 instead of at o-5—50-95—I00, yet the 65
per cent light line cuts the gabbro-curve at rather a high point. The
addition of a fifth class for rocks with approximately equal amounts
of light and dark constituents would increase the total number of
families by 104, and to the writer it seems undesirable to do this.
Not much is gained, and it is just as well to speak of light and dark
gabbros, separating on the 50-50 line, as to make the main gabbro
class the intermediate 35-65 position. In the lowest curve, which
MINERALOGICAL CLASSIFICATION OF IGNEOUS ROCKS 75
represents all rocks, there are no sharp division lines except at 5 and
95 or thereabout. ‘The central division points could equally well
be 50-50 or 35 and 65. On the whole, the writer thinks the 50-50
line best, but leaves this question open for the present.
Lincoln? makes three divisions, leucocratic, mesocratic, and
melanocratic, according to the percentages of light constituents,
with division lines at o-33-67-100; and in the expanded series,
five divisions at 0-4-33-67—96-100.
Iddings? separates his rocks on the ratios o-?—3—100; that is,
into rocks with less than 373 per cent dark, between 374 and 624
per cent, and with more than 623 per cent. This makes the first and
third groups very large. Even the C.I.P.W. general subdivisions
of o-123—-373-625-873-100 would make the first and last groups
too large, for rocks with 123 per cent of dark constituents (see
Fig. 7 with 10 per cent) certainly are not leucocratic. Further-
more, a division at 123 or 373 per cent at the leucocratic end
is not so logical as at 5 per cent (cf. Fig. 12). Shand’ makes his
divisions at 100-97—-go-80-70-60-50-40-30-20-10-3—0 per cent
light minerals. These, however, are too many for the purpose of
classification, the essential difference between rocks with 60 and 70
per cent of dark constituents, for example, being insignificant.
From the curves in Fig. 12 there appears to be little choice between
dividing lines at 33, 35, 373, or 50. If there is any, it is in favor of
50-50.
Orders.—Having divided the rocks into four (or five) classes
according to the amount of dark constituent, they may be divided
into orders on the basis of the plagioclase.
In determining the kind of plagioclase in a rock, it has been
quite customary to give the Ab-An percentage in simple round
numbers, such as Ab,An,, Ab,;An,, etc. This produces an excessive
number of rocks at these points, as is clearly brought out in Fig. 13,
which is less valuable for that reason. As may be seen, there are
crests at Ab, Ab,An,, Ab,An,, Ab;An,, Ab;An,, and AbgAniop.
Having no other marked crests in the curve indicating natural
division lines, the writer has taken the points o-5—50~-95—I100,
t F.C. Lincoln, of. cit., 556.
2 J. P. Iddings, op. cit., II, 150, 308. 3S. J. Shand, op. cit.
76 ALBERT JOHANNSEN
thus grouping albite (allowing up to Ab,;An; for latitude), oligoclase
and andesine, labradorite and bytownite, and anorthite (with
Ab;An,; for latitude), and conforming to the present lines of separa-
tion between the alkali rocks, the acid plagioclase (dioritic) rocks,
the basic plagioclase (gabbroic) rocks, and the anorthite rocks.
Each of the first three classes of rocks may be divided in this
manner into four orders, making twelve orders in all. The fourth
class, that is, the one in which the dark constituents form over 95 per
cent of the rock and the light constituents, including the feldspars,
only 5 per cent, naturally cannot be divided on the basis of the
feldspars; consequently its orders are differently formed.
O15 205220 30 40-50 60 _—-70
80 90 95 100
in loo
a
°
& 10
E
4
AD
Percentage anorthite 4,
Fic. 13.—Number of rocks with various plagioclases, all families from o to 31
included.
Lincoln does not divide his rocks on the kind of plagioclase, but
separates his gabbro from diorite, for example, simply on the basis
of its leucocratic or mesocratic character, which is not according to
common usage.
Iddings' unites his orthoclase with albite and uses the ratio of
orthoclase plus albite to other plagioclases, and makes ae divisions?
at the points o—7—3—3-1-100; that is, at o-123—-373-623-873-100
percent. ‘These divisions are not quite comparable 6 ae present
writer’s triangular divisions into the Kf, Naf, and Caf ratios.
Owing to the fact that soda is of more importance in connection with
the lime of the plagioclases than it is in connection with the potash
of cryptoperthite, it seems more reasonable to separate Kf from
Naf+Caf than to separate Kf+Ab from the plagioclase minus
albite. The latter would be simpler in placing microperthite, but
tJ. P. Iddings, op. cit., II, 41. 2 Tbid., pp. 38, 40-41, 42, 44.
MINERALOGICAL CLASSIFICATION OF IGNEOUS ROCKS 77
is incorrect in theory. Tyrrell* says that Iddings’ system is faulty
in this respect, and suggests uniting all the soda molecules with
orthoclase, and comparing the sum with the lime molecules. But
to this the objection may be made that it fails to separate the soda-
from the potash-rocks. Personally the writer prefers to go one
step farther and separate the three molecules, as shown in
Fig. 11. If Kf and all the Naf were united, it would make
difficulty in the monzonite group where the Ab molecule must
be separated from the An. Thus with the potash and soda
united, a rock with 50 per cent orthoclase and 50 per cent
andesine (Abs«An,) would give (Or+Ab)s,An2o, while *f classified
by the ratio of orthoclase to albite plus anorthite it would
give Or,, Plag;,. The difficulty in determining the albite in most
microperthite is not great; the amount can be estimated with
little error.2 Of course this is not possible in anorthoclase, and
rocks containing much of this mineral will have to be determined
chemically. Ordinarily, however, the amount of soda is too small
to change the classification of the rock, even if neglected. In rocks
which contain known amounts of soda-orthoclase and plagioclase,
the molecules must be separated. Thus a ciminite from the
Roman Comagmatic Region? contains soda-orthoclase (OrsAb,)
43.6 per cent and labradorite (Ab;An,) 16.1 per cent, which
gives orthoclase 37.4 per cent and albite 6.2 per cent from
the soda-orthoclase, and albite 5.4 per cent and anorthite 10.7
per cent from the labradorite. Uniting these there is orthoclase
37-4 per cent, albite 11.6 per cent, and anorthite 10.7 per cent.
This gives Ab,,An,3, the point falling just on the Ab side of Ab,An,
or in Order 2, and Kf,,Plag,,, which brings the rock in the row of
families 3, 8, 13, etc. (Fig. 16). Zonal feldspar may be determined
by considering the approximate amounts of each kind and obtaining
the average Ab-An value. This will be necessary in but few cases,
for ordinarily it may be determined by inspection whether the
™G. W. Tyrrell, “A Review of Igneous Rock Classification,” Science Progress,
No. 33 (July, 1914), 79.
?¥For figures giving a comparison of measured and calculated values see Eero
Makinen, Bull. com. géol. Finlande, No. 35 (1913), p. 74; Charles H. Warren, Proc.
Amer. Acad. Arts and Sciences, LI (1915), 127-54.
3H. S. Washington, op. cit., p. 65.
78 ALBERT JOHANNSEN
average runs across the Ab,,An,, line. Of course if the nucleus
as well as the rim falls entirely between the o-5, 5-50, 50-95, or
95-100 lines, there is no need for computation unless it be to
determine the exact position of the rock in the triangle.
Families —The quarfeloid face of the double tetrahedron
(Fig. 9), or any face parallel to it, will appear as shown in Fig. 16.
To locate the lines separating the various families it was necessary
to determine the logical divisions in two directions; namely, between
a Se anaes ee ela
Wed ia SP a
EE
cy Ee ee ee
eee eee eee
2 ee ee
ee Se ee
ee RC Os
Ne Zann
BOeran= = aeraor ee
i
Pla ) TOMASO 30 ie “s 90 0
Kf 100 90 80 70 60 20 45 30 10 a)
Percentage plagioclase
Fic. 14.—Ratios of Kf to plagioclase in Families 1 to 15, Orders 1 to 3, and totals.
Vertical scale of totals is one-half of other curves. The numbers indicate the orders.
the potash feldspar and the plagioclase and between rocks with or
without quartz or feldspathoids.
In Fig. 14 the curves for the proportions of potash-feldspar to
plagioclase are shown for Orders 1, 2, and 3 and for the sum of all
feldspathoid-free rocks; the writer having no mode-analyses showing
potash-feldspar with anorthite in Order 4. The curves show
rather excessive increases on the 50-50 line, due to the fact that
many writers speak of labradorite as Ab;An,; the deduction of
rocks where this was done would slightly reduce the lines. In all
the curves the dividing lines may be made at o-5-35-65-95-I00,
corresponding to the subdivisions in vogue of alkali-granite,
granite, quartz-mionzonite, granodiorite, quartz-diorite, etc.
MINERALOGICAL CLASSIFICATION OF IGNEOUS ROCKS ‘79
The vertical direction of Fig. 16 gives the quartz percentage. In
Fig. 15 are plotted the curves for the proportion of quartz among
the light constituents for all rocks in the upper triangle (Families
© to 15, Fig. 16), and separate curves for Orders 1, 2, and 3. The
separation at 5 is clear. There may be a question whether the
upper division of quartz should be made at 95, go, or even at 65.
For symmetry, of course, it should be at 95. With respect to a line
at 50, the writer is in doubt. Practically all the rocks fall below
50 per cent quartz (that is, quartz is less than 50 per cent of the
et Op of quartz
100 90 60 50 40 30 20 10 10
ee TU Ca Ee a
eal Le Se Pe
Si Ce ee eee
=. ee ee
Sealab
Biol | ee
E ee
eS Se
Cea eae
SS as
Fic. 15.—Percentage of quartz among the light constituents, recalculated to roo.
Curves for Orders 1, 2, 3, 4, and totals. The numbers indicate the orders.
light constituents, consequently it forms even less than 50 per
cent of all the constituents of the rock). It would be possible to
group all the rocks given in Families 2 and 7, 3 and 8, 4 and 9, etc.
(Fig. 16), together, and call those falling in the upper divisions
simply quartz-rich granites, etc. However, since there are so few
rocks here, it may make it all the more desirable to divide on the
50-50 line. This would make uniform divisions everywhere
_ in the system at o-5—50-g5—100 except for the Kf-Plag ratio. Of
course the retention of the line at 50 in this and the lower triangle
makes 8 or ro more families in each order of the first three classes,
or a total of 102. However, if these families are simply numbered
and the rocks called quartz-rich granite, quartz-rich granodiorite,
nephelite-rich nephelite-syenite, etc., it will add no new names and
make clearer the positions of the rocks. Curves drawn for the
80 ALBERT JOHANNSEN
feldspathoid rocks are similar to those in Fig. 15, but are somewhat
more irregular owing to insufficient data.
The families are to be numbered as shown in Fig. 16. The
Kf
Foide
Fic. 16.—Family numbers in Classes
I to 3.
object in beginning with o is to make the positions easier to remem-
ber, since they run in groups of five. Furthermore, Family o
occurs only in Order 1, as do also Families 1, 6, 11, 16, 21, 26, and
31, for they form the hinge about which the order tetrahedron
(Fig. 10) was opened, and are
the same inall. This is shown
in Figs. 21 to 23, where these
families are omitted and repre-
sented by dotted lines. Instead
of having 12X 32families, there-
fore, there are 3 X 32 families (in
Ta ws es \ the first orders in each of the
3 [19 740 Pleg first three classes) +9 X 24 fami-
lies (in Orders 2, 3, and 4) +3
X15+1 families (in Class 4, to
be mentioned later), making
358 families in all. If Order 1
is omitted, as suggested in ques-
tion 4, below, the total families
will be 286, and if Order 4 is
united with Order 3 there will
be only 214. Although the
maximum number of families is
358, it does not mean that there are 358 names to learn, for the
light and dark rocks may be separated by prefixes without making
awkward names; thus leuco-granite, melano-granite, etc.
The divisions made by other writers may now be compared
with Figs. 14 and 15. Lincoln uses the ratio orthoclase to all
plagioclase, the latter not differentiated as is done here. His per-
centages are 100-96-67—33-4-0.
It is rather difficult to compare the divisions proposed by Iddings
with those proposed by Lincoln or by the present writer, for, as
mentioned above, he unites albite with the potash feldspar and
MINERALOGICAL CLASSIFICATION OF IGNEOUS ROCKS 81
compares this sum with the remaining plagioclase; that is, he has
the ratio
Kf+-Ab in albite
Ab in soda-lime feldspar-+-all An °
His divisions are,’ as mentioned under “Orders,” above, 100-873—
623-379-125-0.
Olivine
Pyroxenes
Fic. 17
Fic. 17.—Subdivisions of the tetrahedron of Class 4 into orders
Fic. 18.—Subdivisions of Orders 1 and 2, Class 4, into 15 families. Order 3 is
subdivided similarly, but the corners represent olivine, biotite and amphiboles,
pyroxenes, and the ‘‘ores.” Order 4 has the various “‘ores’”’ for corners.
Amphibole, Olivine Amphibole,
6 / t. 12 /Blotite
Ne is
__10 \orthorh.
/ pyroxene
\ ye
Monoch./ 8
os
pyroxene,
\
Malioies
Me
Amphibole,
Biotite
Fic. 19.—Family numbers in Class 4
The quartz- (or feldspathoid-) feldspar relations given by Lincoln
are 100-96-67—33-4-0, and by Iddings? 100-623—123—0. Lincoln’s
division at 33 does not fit at all well into Fig. 15. Idding’s divisions
tJ. P. Iddings, op. cit., II, 38, 40-41, 42, 44. 2 [bid., pp. 32, 38, 147, 228, 202.
82 ALBERT JOHANNSEN
fit quite as well as the divisions 100-95—50-5-o proposed in the
present paper, but the writer feels that a rock with 123 per cent
quartz (see Fig. 7 with ro per cent) is too rich in quartz to be called
a syenite. The writer would have no objection to making the
divisions at 100-95—65~5-o quartz (or feldspathoids), that is, on
the basis of the 100-95—56—35—5—o divisions with the omission of the
25 \per cent: line but
thinks it better to
leave the divisions
symmetrical. A rock
with over 50 per cent
quartz or feldspathoid
is certainly distinct
enough to deserve a
separate place.
Class 4.—Owing to
the absence of light
{7 NY: constituents in Class 4
ee eee "it was necessary to
WWAATIAARALY
Ss MBO
INT NUS
LSASAAVANN
LENS
make the subdivisions
on a different basis.
After numerous at-
tempts with different
figures and different
groupings of minerals,
it was found that the
compartments shown
in Fig. 17 correspond
most closely to. the)
present subdivisions of
Fic. 20.—Rocks of Order 1 falling in Classes T to13. the melanocratic rocks
Open circles are rocks of Class 1, dark circles rocks of é
Class 2, and triangles rocks of Class 3. The tetrahedron is sub-
Lo
Foids
divided into fourorders |
by planes parallel to the left-hand face, each order represent-
ing an increasing amount of the ores. The division points for
these planes, as in the other classes, are o-5—50-95—100. To
accommodate the rocks of the old classification, each order triangle
MINERALOGICAL CLASSIFICATION OF IGNEOUS ROCKS 83
was opened out at one corner into a secondary tetrahedron, as
shown in Fig.18. ‘The division points between families are at o—-25-
75-100 to make the nomenclature conform to the older systems, and
they are numbered, from the top and counterclockwise, from 1 to
15. The four corners, in Orders 1 and 2, represent respectively oli-
vine, biotite and amphibole, monoclinic pyroxene, and ortho-
rhombic pyroxene. au
In Order 3 the corners
are olivine, biotite
and amphibole, the
pyroxenes, and the
Hones) and, other
dark constituents. In
Order 4, if thought
desirable, they may y, AN
NZ,
ee
a fk foe iN
‘ NENTS “fa =
the writer, however,
groups the ores in one \
family, for, considered
as rocks, they are un-
important and hardly
worth while separat-
ing. All of the fam-
ilies of the whole class,
except Family 15, ap-
pear on the surface
of the tetrahedron,
Families 5 and 11.
being at the back of
Fig. 18, Family 14
underneath, and Family 15 in the center. Fig. 19 shows the
tetrahedron opened out; Family 15 alone not appearing.
Coy
QWOASHOY
Vv
Foids
Fic. 21.—Rocks of Order 2 falling in Classes 1 to 3
ROCKS INCLUDED IN THE VARIOUS FAMILIES
Computed by the rules which follow, nearly 600 rocks are
represented in Figs. 20 to 23. In these diagrams the rocks of the
84 ALBERT JOHANNSEN
same order, though of different classes, are shown together, the
leucocratic rocks of Class 1 being represented by open circles,
the moderately dark rocks of Class 2 by dark circles, and the
dark rocks of Class 3 by triangles. The larger circles and triangles
indicate that a number of mode-analyses fall together at these
points. It will be seen that there are 32 families represented in
Fig. 20, while in the
other three figures
there are only 24 to
the order, as_ ex-
plained above.
In the following
list about 500 com-
puted rocks are ar-
ranged according to
their old names fol-
lowed by numbers
aw Goes ese SOEATAD Ge indicating their posi-
\. ieee ere eariry psec" tions in the present
classification. No
FOCKS,) are given
having less than three
mode-analyses unless
of well-defined recent
rocks. The first fig-
ure in the following
numbers represents
the class, the second
y the order, and _ the
Fic. 22.—Rocks of ae falling in Classes 1 to 3 third (or aur an
fourth) the family.
There are no orders in Families, 0, 1, 6, 11, 16, 21, 26, and 31,
but since the rocks of these families are plotted in the double
triangles of Order 1, their positions may be indicated by the
figure 1. ‘The figures in parentheses indicate the number of deter-
mined rocks which fell into that family. For example, 2123, 118,
422, 4210 represent respectively Class 2, Order 1, Family 23;
ANYON
MINERALOGICAL CLASSIFICATION OF IGNEOUS ROCKS 85
Class 1, Order 1, Family 8; Class 4, Order 2, Family 2; and
Class 4, Order 2, Family 10. Numbers in bold-face type in the
list indicate proper families among scattered (misnamed) rocks.
The rocks may seem rather scattered, but upon careful com-
parison it will be seen that they are not so far apart as they appear
at first sight, for the lighter and darker rocks of Classes 2 and 3
in the gabbro and
diorite families and of
Classes 1 and 2 in
the granite, syenite,
monzonite, etc., fam-
ilies have ordinarily
not been separated in
petrographic reports.
Much of the variation,
of course, is due to
the loose naming of
rocks, and an andesine-
bearing rock with
augite may have been
named gabbro, and a
mesocratic labradorite-
augite rock, diorite, etc.
Thus the basalts in the
list below—2315(4),
3215(1), 3315(5)—in-
clude g rocks belonging
to Order 3, Family 15,
the normal gabbro-
basalt family, 4 of the
rocks being mesocratic
and 5 melanocratic-
ANIA
ERY
Foids
Fic. 23.—Rocks of Order 4 falling in Classes 1 to 3
The remaining rock (3215) is a melanocratic andesine-bearing
rock which should have been called an andesite.
Andesite 2215(6), 3215(1).
Aplite 128(3), 217(1), 223(1).
Alaskite 118(2), 117(2), 128(1).
86
ALBERT JOHANNSEN
Basalt 2315(4), 3215(1), 3315(5).
Basalt, Quartz 2310(3).
Bostonite 2114(1), 2112(1), 2214(1).
Camptonite 3215(2), 3214(1).
Comendite 217(1).
Covite 2123(1).
Diabase 2215(1), 2315(9), 3215(1), 3315(4).
Diorite 2214(2), 2215(4), 3215(2).
Diorite, Quartz 2210(6), 238(1), 239(4), 2310(3).
Essexite 2320(1), 2315(1), 2324(2), 3213(1), 3314(2).
Gabbro 2314(1), 2315(14), 3314(2), 3315(5)._
Gabbro, Quartz 3310(3).
Gauteite 2320(1), 2330(1).
Granite, including alkali-granite 117(4), 118(1), 123 (1), 127(3), 128(2),
211(1), 212(4), 217(20), 218(8), 210(1), 227(21), 228(27), 229 (16),
2210(6), 238(1), 2310(1).
Granite-porphyry 212(1), 216(1), 227(2).
Granodiorite 228(2), 229(8).
Grorudite 218(6), 219(2).
Hedrumite 2123(1), 2114(1), 2124(1).
Hornblendite 4212(1), 4112(1).
Heumite 2223(2), 2224(2).
Tjolite 2131(5).
Kersantite 2214(1), 3215(2), 3315(1).
Leucitite 2131(1), 2220(1), 2230(1), 2329(1), 2430(2), 3430(1).
Lindoite 228(5).
Laurdalite 2124(1), 2224(1).
Laurvikite 2118(1), 2123(z).
Leucite-tephrite 2223(5), 2224(4), 2220(3), 2327(1), 2230(2).
Malchite 3210 (3).
Melilite-basalt 2315(3).
Minette 216(2), 2t11(1), 227(1), 2212(z), 2215(1).
Minette, Soda 2114(1), 220(1), 2214(2).
Mariupolite 2125(2).
Missourite 3131(3).
Monmouthite 2131(1).
Monzonite 2213(5), 2313(1), 3213(1), ey
Monzonite, Quartz 128(4), 1290(1), 228(7), 220(4), 2214(1), 238(3), 322(3).
Nephelite-syenite 1224(1), 2122(3), 2123(3), 2124(1), 2126(1), 2129 (1),
2222(3), 2223(1), 2225(1).
Norite 2214(1), 2314(3), 3214(1), 3314(5), 3315(5).
Pantellerite 218(3).
Pegmatite 117(7), 118(2), 127(1), 129(4), 1210(2).
Phonolite 2122(2).
MINERALOGICAL CLASSIFICATION OF IGNEOUS ROCKS 87
Rockallite 215(2).
Rougemontite 2415(1).
Rouvillite 2225(1).
Shonkinite 2112(1), 3112(1), 3212(1).
Solvsbergite 2112(2), 2113(4), 2123(2).
Syenite’ 2111(2), 2113(3), 227(11), 228(7), 2209(1), 22¥2(2), 2312 (1),
2313(1), 3212(1), 327(1).
Tawite 2127(1).
Tinguaite 2122(1), 2123(4), 2124(3), 2116(1).
-Trachyte 216(1), 2113(1), 2123(1), 2212(1), 2213(2).
Vulsinite 2213(1), 2222(1), 2223(2).
Yamaskite 3415(3).
CLASS NAMES
In a few cases the old classifications give special names to the
dark varieties of feldspathic rocks. ‘Thus shonkinite was definitely
defined as a syenite with more than half of the constituents dark,
although in the foregoing list one rock (2112) is mesocratic. In
most cases, however, there are no special names for the dark
feldspathic rocks, nor is it necessary to invent such, for the differ-
ent varieties may be distinguished by prefixes. Since the rocks of
Class 4 are separated from each other on an entirely different basis
from the rocks of the other three classes and have special names
they need not be considered here. ‘To the other three classes the
names suggested by Brégger—leucocratic, mesocratic, and melano-
cratic—may be prefixed. If desired, a rock may be called a leuco-
granite, meso-granite, or melano-granite, for example, instead of a
leucocratic granite, mesocratic granite, etc. Meso, unfortunately,
has been used as a prefix for Mesozoic rocks, but since the age
classification of igneous rocks is no longer in use this would cause
no confusion. Furthermore, since the normal rock usually falls
in Class 2, the meso prefix is seldom necessary, and its name may
be used without a prefix.
ORDER NAMES
The different orders may be indicated, when no special names
exist for the various rocks, by the prefixes albite- (or soda-), sodic-,
calcic-, and anorthite- (or lime-). Thus in the diorite family the
rocks of the different orders would be albite- (or soda-) diorite,
a
88
ALBERT JOHANNSEN
ov
3
SF
=
oO e
a =
3 a
rn v
=e a
3 2
a
Potash- = eaamatontbeanes lorite
ayer te Alk.-sonz Alk.-syenodior\ Nat Ke g CaNat
ee
V
Folds
Fic. 24 Fic. 25
“qu
Rr ¢
Lherzolite
Saxonite
Augitite Enstatolite
Diall agite Hypersthenite
Websterite
Orthorh.
pyroxene
pyroxene
FIG. 27
Foids
Fic. 26
Fic. 24.—Family names, Class 2, Order 1
Fic. 25.—Family names, Class 2, Order 2
Fic. 26.—Family names, Class 2, Order 3
Fic. 27.—Family names, Class 4, Order 1, Families 1, 2, 3, 4, 8, 9, and 10
1
MINERALOGICAL CLASSIFICATION OF IGNEOUS ROCKS 89
sodic-diorite, calcic-diorite, and anorthite- (or lime-) diorite. As
a matter of fact, these rocks in the old classification have special
names, namely, soda-syenite, diorite, gabbro, and anorthite-gabbro,
and, except the first, which more properly is an albite- (or soda-)
diorite, should not be changed. The prefixes persodic, dosodic,
etc., of the C.I.P.W. system cannot be used, since they apply to
definite proportions of the constituents and not to those used here.
FAMILY NAMES
It is not the intention in this paper to name definitely all the
families, those in Figs. 24 to 27 being given simply as examples.
Most of the family names have been determined, and will be given in
a succeeding paper. The family name should be that of a rock with-
out abnormal constituents which occupies nearly the center-point of
that family. Thus a garnet-bearing rock should not be chosen as a
family representative if a non-garnetiferous rock is known, the
garnetiferous rock being indicated by a prefix. The name should
also be that of the plutonic rock, if such is known. Furthermore,
if only one name is given to the rocks of the same family in the
various classes, it should be given to Class 2; Class 1 will then be
its leucocratic variety and Class 3 its melanocratic variety. It is
not to be understood from this that the writer thinks it undesirable
to name particular varieties, for it may be very desirable if they
represent distinct types and if their relationships to known rocks
are clearly shown; but if a new type differs only by the presence
of a single abnormal constituent, that constituent should simply
be used as a modifying name.
The reasons for using certain family names, such as adamellite
for quartz-monzonite, tonalite for quartz-diorite, etc., will be given
in a succeeding paper. Syenodiorite, syenogabbro, and grano-
gabbro are introduced as new terms to fill definite positions, the
last being the orthoclase-bearing variety of quartz-gabbro and
analogous to granodiorite, the first two being the quartz-free
varieties of granodiorite and granogabbro.
Sub-families in Orders 1, 2, and 3 are formed on the basis of the
predominating dark or auxiliary constituent; thus under granite
are the divisions biotite-granite, hornblende-granite, topaz-granite,
go ALBERT JOHANNSEN
tourmaline-granite, etc. This applies whether the modifying
constituent is a mafite or an auxiliary.
THE MINERAL GROUPS
It is not sufficient to divide the constituents of the rock into
those that are light and those that are dark, but it is necessary to
make certain definite groupings. The primary division, of course,
is into quarfeloids and mafites. Under the former are included:
QUARFELOIDS
Quartz (Qu).
Potash feldspar (Kf), including orthoclase and microcline, and the ortho-
clase molecule in microperthite, anorthoclase, etc.
Plagioclase (Plag), including the albite molecule in anorthoclase as well as
all plagioclases.
Feldspathoids (Foids), nephelite, leucite, sodalite, hauynite, noselite,
melilite, primary analcite, primary cancrinite, eudialyte, etc.
The rear angle of the double tetrahedron represents the mafites.
It is the position of the remainder after the quarfeloids and
auxiliary constituents have been deducted.
MAFITES
Dark micas (biotite, phlogopite, etc.). !
Amphiboles.
Pyroxenes (including uralitized pyroxenes).
Olivine.
Iron ores (magnetite, ilmenite, chromite, pyrite, hematite, etc.).
Cassiterite.
Garnet.
Primary epidote.
Allanite, zircon, rutile, and other dark minor accessories.
SECONDARY CONSTITUENTS
Secondary constituents are calculated as the originals from
which they came. Thus ore replacements of the mafites are com-
puted as mafites, kaolin as feldspar, chlorite as a biopyribole,
cancrinite and analcite as feldspathoids, serpentine as a mafite, etc.
MINERALOGICAL CLASSIFICATION OF IGNEOUS ROCKS QI!
AUXILIARY CONSTITUENTS
Auxiliary constituents are constituents, mostly pneumatolytic
or metamorphic, which may be used in the nomenclature as min-
eral modifiers in the formation of sub-families. Rocks containing
these minerals may have independent names if desired. The
auxiliary minerals are seldom of importance.
Topaz Primary scapolite
Tourmaline Muscovite
Cordierite Lepidolite
Corundum Zinnwaldite
Fluorite Apatite, etc.
Andalusite
It will be observed that most of the auxiliary constituents are
light in color; they are, consequently, computed among the leuco-
crates. It is true that if this is done, tourmaline-granite will fall
among the leucocratic rocks, but since this rock is aplitic and the
mineral pneumatolytic, this is not undesirable.
Glass must be computed from an analysis. One can usually
surmise its composition from the character of the phenocrysts and
the appearance of the rock. When undetermined, the rock must
be given a tentative name, such as hyaline-rhyolite, etc.
RULES FOR COMPUTING ROCKS FROM THEIR MODES
1. The sum of the minerals in the mode should be 1oo+o.5.
Ii less, recalculate? to roo. The sum of the leucocrates (quarfe-
loids plus auxiliary minerals) so obtained determines the class.
Class 1. Leucocrates form less than 95 per cent of the total rock.
Class 2. Leucocrates between 95 and 50 per cent.?
Class 3. Leucocrates between 50 and 5 per cent.
Class 4. Leucocrates less than 5 per cent.
2. Determine the orders in Classes 1, 2, and 3 directly from the
Ab-An ratio, the division lines being o-5—50-95—-100. In rocks
t All of the necessary computations may be performed in an instant of time by
means of a slide-rule.
2 These classes are tentative. If thought desirable (see question 1, below), the
rocks will be divided into five classes.
Qg2 ALBERT JOHANNSEN
containing both anorthoclase and soda-lime feldspars, the three
molecules Kf, Naf, and Caf are to be separated, and the orders
determined by the total Ab-An ratio. (See above, under the
heading ‘‘Orders,”’ for an example of ciminite so separated.) In
Class 4 the orders are determined by the percentage of “‘ores’’ and
other dark minerals in the rock, the division points also being
O-5—50—-95—100.
3. Determine the family. In Classes 1, 2, and 3 first recalculate
the quarfeloids to 100. The amount of quartz (or feldspathoid)
immediately determines the distance from the feldspar line. The
separation points are o-5—50-95-100. Now recalculatet the Kf
plus plagioclase to 100, and determine the proper point on the Kf-
Plag line. (If plotted graphically, the family is directly determined
by the position of the intersection of the three lines. If the
point falls very close to a division line, it may be necessary to
compute its position accurately.) The separation points for Kf-
Plag are o—5—35—65—95—100.
In Class 4, Orders 1 and 2, recalculate the olivine, pyroxenes,
biotite, and amphiboles to too and find the proper positions graphi-
cally, or find the position analytically by taking the ratio of the
minerals of one corner to each of the others; thus augite to olivine,
augite to hypersthene, and augite to biotite or amphibole. The
division points are o-25—75-100. In Class 4, Order 3, the corners
represent olivine, amphibole and biotite, all pyroxenes, and the
“ores” and other dark constituents. In Class 4, Order 4, the
writer groups all the ores in a single family, but classifies the vari-
ous hematite, ilmenite, magnetite, etc., ores as subfamilies. If
desired they may be further separated. If accessory dark min-
erals, not used in the computation, are abundant, they determine
subfamilies and may be mentioned in the rock name.
A few points to be observed——Any percentage value falling
exactly on a line should be moved in the direction of the center of
the triangle. Thus a syenite with 5 per cent quartz is classified
with granite, a rock with 95 per cent mafites belongs to Class 3,
t It is immaterial whether the orthoclase-plagioclase ratio is taken from the original
values or from those reduced as quarfeloids to 100. The results are naturally the
same.
MINERALOGICAL CLASSIFICATION OF IGNEOUS ROCKS 93
and one with 95 per cent quarfeloids to Class 2; Ab,;An; belongs
to Order 2 and Ab,An,,; to Order 3. If the divisions fall on the 50-50
line of quartz they are moved upward, or, with the Foids downward,
toward the apex; that is, they are placed in Families 1 to 5 or 25
to 30. Along the plagioclase line, Ab,,An, is classed with the basic
plagioclase, and 50-50 light-dark with the dark. Rocks falling
on the line separating the two triangles, namely, on the feldspar
line, should be classed on the quartz side, that is, on the normal side.
EXAMPLES
Example 1.—A granodiorite having the composition
Ouartge wee Aves ev wane al 18.0 = 23.1
Orthoclase enue pew eei as} 18.0 = 23.1
Andesine;(AnzovANss))..-). 2... 42.0 = 53.8
Total quarfeloids;.):..:... 78.0
IBIOLITC ME el eenec ete tausaluisS 12.8
Hornblendee iO .sucasue seat, 9.0
IMA CME LICH eta cata Wena. I
STM GAMNT SRI i ean Lente I
otallomaatitesw sq sineuss mice cu. 22.0
100.0
Percentage quarfeloids=78. Rock belongs to Class 2.
AbyAnyo falls between 95 and 50. The order, therefore, is 2.
The family may be rapidly determined graphically, Plot
23.1 Qu, 23.1 Or, and 53.8 CaNaf by measuring 23.1 upward from
the base of the triangle toward Qu, and 23.1 from the right-hand
inclined line toward the lower left corner. The intersection of the
two lines will fall in Family 9 and determines the position of the
rock. Asa check, the point must also lie 53.8 from the left sloping
line toward the lower right corner.
To compute the family analytically: From the presence of 23.1
per cent quartz, the family must lie between numbers 6 and tro, since
there is more than 5 per cent and less than 50 per cent quartz.
Or 18 _ 30
CaNat 42 70°
between 5 and 35 per cent, the family belongs in No. 9.
The rock number, therefore, is 229, that is, Class 2, Order 2,
Family 9
Further, the ratio and since the orthoclase is
94 ALBERT JOHANNSEN
Example 2.—A syenite having
CECE as ira RS ee nr areee arene 60.0 = 76.0
NewS a itieigas sey etle dR Ts, aye ts 18.0 = 22.8
LOH RU ie Uae scans yet ie 1.0 =- 1.2
Total quarfeloids........ 79.0 100.0
BIO tis eh eye ope rete 18.0
oY Derren eee RN GR arena a 20
NIC COS ihe ital unt tay eave I
Motal mafhitesaosn4 ween. 21.0
I00.0
Percentage quarfeloids to mafites 79, therefore Class 2.
Ab,An,=Abg,.;An,,.;, therefore Order 2.
Quartz less than 5 per cent, therefore between Families 11
and 15.
Ki 100077 ; .
CuvaiT i loa: therefore Family 12. The rock number is
2212; thatis, Class 2, Order 2, Family 12. The values 76, 22.8, and
1.2 are used in the graphical location of the rock.
Example 3.—A nephelite-syenite with
| UA Re tea saree ants TESS, fe 205 = 39.0
Nai sre aie. ayia BRIBE)
Ab oA =
‘ ne Oe No eee 2 i ene
Total teldspar s 3 33949: 55.0 100.0
Nephi so Aan hohe pee
Sodalltvis heute seas eee Sas
Total feldspathoids ...... 36.0
Total quarfeloids ....... QI.0
COSA UI ye aires ulege encom Mas 5.0
BIO EA erica clare selec icteetele vans 2}
A CCOSUAR TS Faroe i oho ate ean Lely)
Motalimafites sme sre: 9.0
100.0
Quarfeloid ratio 91. Class 2.
Alga Alig 3 vette Order 2.
36 39-5
= . Between Families 21 and 25.
5 OONS
Foids to feldspars=
MINERALOGICAL CLASSIFICATION OF IGNEOUS ROCKS 95
Kn 215) 39.0
(CHIN GED AoW) Cpeste) |
Rock number is 2223. The values 39 and 61 are used in
plotting the rock.
Family 23.
Example 4—A l|herzolite with
BANE STE GS suena Ulta Loe 45.0 = 47.4
Eby Perstheneni isn cine se ot, 20.0 = 21.0
Olivas WU ues Soe 2) 30.0 = 31.6
iEfornblend eerie. bene, 31,0111.
Miao MeLILCH Rt mia nares Santen. AMO)
100.0
Since there are neither feldspars, feldspathoids, nor quartz, the
rock must belong in Class 4.
The ratio of ferromagnesian minerals to ores is 98 : 2, therefore
the Order is 1.
The ratio of augite to hypersthene is 45: 20= 69:31, therefore the
family lies in the middle row and is either 1, 3, 9, or 15 (Fig. 109).
The ratio of augite to olivine is 45:30=60: 40, and the rock again
lies in the middle line including Families 2, 3, 10, and 15. The
ratio of augite to hornblende is 45:3=95:6, therefore it is in
the front series of families including 1, 2, 3, 4, 8,9, 10. Family 3 is
the only one common to the three computations, consequently the
rock number is 413.
Graphically the rock may be plotted by using the numbers 47.4,
on Owand 22.0.
One of the advantages of this system of classification is that
each thin section of the rock may be plotted independently; the
center point of all the dots representing sections from a single rock-
mass will represent the average. This is much more satisfactory
than estimating the average from a number of sections which differ
considerably in the amounts of the constituents. The various
dots representing complementary rocks will fall in straight or
branching lines, showing the course of differentiation.
Before publishing his second paper on this system of classi-
fication the writer desires the opinions of more petrographers than
he has been able to consult personally. He would be very glad,
96 ALBERT JOHANNSEN
therefore, to receive at once answers to the following questions as well
as further comments from all who are interested.
QUESTIONS
1. Classes.—Should there be a fifth class for rocks having
approximately equal amounts of light and dark constituents? The
limits would then be o-5-35—65—95—I100 instead of o-5—50-95-I00,
as here proposed. The introduction of an extra class would add
104 families.
2. Orders.—Should Order 4 (Fig. 23), in which there are very
few rocks, be combined with Order 3? Order 3 would then contain
all rocks with plagioclase from labradorite to anorthite inclusive.
This would make the subdivisions from Ab to An at o-5—50~-100,
and would reduce the number of families by 72. Of course, if the
fourth order is retained the pigeonholes need not be named until
rocks occupying them have been found.
3. The line separating the granites, adamellites, etc., from the
corresponding quartz-rich varieties is here taken at 50 per cent
quartz. Should there be a dividing line here, or should granite, for
example, include all rocks having from 5 to 95 per cent of quartz ?
As suggested above, the division line might be made at 65, making
the lines o—5—65—95—100.
4. In the older classifications albite is united with orthoclase
for the alkali rocks. This would throw out Order 1, but in the
older systems, with the introduction of lime, the soda molecules are
divided into two parts, and orthoclase plus albite is contrasted with
the lime-soda plagioclases. This division is not logical, but is it
desirable? Ifsucha division were made, Order 1 (Fig. 20) would be
dropped and the alkali rocks would form Families 1, 6, 11, 16, 21,
and 26 of the triangles now representing Order 2 (Fig. 21), and
soda- and potash-rocks would have to be separated in the sub-
families. The double triangle would then have orthoclase-+albite
-+microperthite+ anorthoclase for the left angle of the base, while
the right corner would be CaNaf, NaCaf, or Caf, depending upon
the orders. Such a combination would simplify the placing of
rocks containing microperthite, which is worth careful consideration,
but the grouping is not so correct theoretically. All of the rocks
MINERALOGICAL CLASSIFICATION OF IGNEOUS ROCKS 97
of Fig. 20 would then fall into the dotted compartments of
Fig. 21. Computed modes, however, would be more difficult to
place. As a matter of fact it is usually not difficult to separate
the albite in microperthite from the orthoclase. Should this change
be made, Family 6, for example, would become the family of the
alkali-granites, and would contain potash-granite, alkali-granite,
alkali-adamellite, alkali-granodiorite, and soda-tonalite. The latter
would then again become soda-granite, the first potash-granite,
and the intermediate rocks soda-potash granites. Covite, mariupo-
lite, most essexites, etc., would fall in Family: 21 without differentia-
tion. Such a combination would reduce the number of families by
72, and if the anorthite were united in Class 3, as suggested above,
the total reduction would be 144 families. Personally the writer
is inclined to favor separating the feldspars into the Or, Ab, and An
molecules.
5. Would it be desirable to indicate, in the name of the rock
itself, that the mineral proportions have been determined, and that
the rock falls into a certain compartment, for example by a slight
change in the spelling, such as granyt, dioryt, etc.? Of course
terms like monzonite BRrOGGER, theralite ROSENBUSCH, etc.,
might be used, but they seem cumbersome. (Granyte, dioryte,
etc., cannot be used, since this spelling was suggested and used by
Dana to contrast with the -zte endings of minerals.)
A ppendix.—An alternative classification could be based upon
four double tetrahedrons, representing four classes, according to
the amounts of light and dark constituents, and each subdivided
as in Fig. 27. The corners of the tetrahedrons would be quartz,
Kf, Naf, Caf, and Foids, and the division points o-25—75—100.
There would be fewer varieties than in the preceding classification,
and it would be much simpler, but the families would not corre-
spond so closely to those in the old classifications as does the one
given above.
REVIEWS
Geology of the Hanagita-Bremner Region of Alaska. By F. H.
Morrir. U.S. Geol. Survey, Bull. No. 576. Pp. 55, figs. 6,
pls. 6, maps 2.
The area described in this report is in the southern part of the Copper
River drainage basin. Chitina River bounds it on the north, and it
extends southward half-way to the coast.
Field work in this region was of a reconnaissance character, but the
larger stratigraphic units have been outlined. The oldest sediments
are mainly schists, slates, and limestones, and have been referred to the
Carboniferous. These beds have been deformed by close folding and
faulting and cut locally by intrusions. Unconformable above them is
a series of interstratified beds of slate and graywacke thought to be
equivalent to the Valdez series, and early Mesozoic in age. This series
is in turn unconformable beneath conglomerates and tuffaceous slates
of Middle Jurassic age.
The district presents a number of problems in physiography. The
drainage has a rectilinear arrangement which must bear some close rela-
tion to geologic structure. All the valleys have been profoundly glaci-
ated. Many streams are now eroding valley trains. A number of
situations appear very favorable for stream capture.
The author is inclined to doubt the theory that Copper River is an
antecedent stream across the Chucagh Mountains. He suggests that
ice erosion over a narrow divide enabled a southward-flowing stream to
tap the Copper River and divert it from a westward course. To com-
plete this theory it seems necessary to assume uplift along the western
part of the basin to check the flow in that direction, and that along a
great part of its course the Copper River has been reversed since the
retreat of the ice. W. Bowe
The Shinumo Quadrangle. By L. F. Nosie. U.S. Geol. Survey,
Bull Nows46o.7 (Pp stco gig nr splices.
The remarkable geologic section exposed in the Shinumo quadrangle
rivals those that have been described previously in the Grand Canyon. -
The generally unaltered condition of the beds, the great vertical extent
98
REVIEWS 99
of the exposures, and the absence of a vegetal mask reveal the geologic
history in great detail.
The rocks in the quadrangle range from Archean to late Paleozoic in
age. The pre-Cambrian portion of the section follows:
Proterozoic
Grand Canon series (Unkar group)
Great unconformity
Dox 'sandstoneree swe ata cen ke oa 2,297 feet.
Shimumoquartzite™. .)33s 10s. see TS OAGIE vs
takcataishale aces ne ecole oes Sins SO. e:
Basswlimestone nie. mc naies st cas ict aa Baca
HO tautay COMO sities hese (ets sie neue OO;O! a
Archeozoic
Great unconformity
Vishnu schist
The Proterozoic sediments were deposited on a surface that repre-
sented almost perfect peneplanation. At the close of the period of
deposition, uplift and great normal faulting inset these beds deeply into
the Archean. This led to their preservation during the next period
of great erosion, which again resulted in peneplanation by the close of the
pre-Cambrian. Where not protected by faulting the Proterozoic beds
were removed. The remnants are in great wedge-shaped masses, each
bounded by a fault plane, and the two great erosion surfaces. In no
other known region do two profound peneplains meet in a line.
Cambrian and Carboniferous sediments exist throughout the quad-
rangle. A disconformity represents the intervening systems. Mesozoic
and Tertiary rocks ranging up to 6,000 feet in thickness formerly covered
this-area. In early Quaternary times a cycle of erosion, known as the
“reat denudation,” drove their outcrops many miles to the north.
The writer follows Davis and others in recognizing but two cycles
of erosion in the formation of present physiographic features. The first,
the great denudation, developed a virtual peneplain, and the second,
during the latter part of the Quaternary, resulted in cutting the Grand
Canyon. The Esplanade and Tonto platforms, explained by Dutton as
temporary base-levels, are held to be structural benches.
The writer also follows Davis in holding that the present course of
the Colorado River was established before the beginning of the uplift
-that resulted in the canyon cycle of erosion. It is a superposed stream,
let down from the surface of the peneplain of the great denudation.
W. B. W.
TOO REVIEWS
Gypsum Deposits of the Maritime Provinces. By Wititam F.
JENNISON. Canada Department of Mines, No. 84, tort.
Pp. 170, figs. 19, pls. 36.
This report is largely taken up with general discussion of the world-
distribution of gypsum, its origin, manufacturing processes, arid the
character of the manufactured products. Considerable space is given
to descriptions of various local occurrences that may become of com-
mercial importance.
Nova Scotia, New Brunswick, and the Magdalen Islands make up
the Maritime Provinces. The gypsum deposits were thought at one
time to belong to Permian age, but they are now known to be Mississip-
pian. In Nova Scotia the deposits are not limited to any particular
horizon, but are found near the base, in the middle of the system, and
immediately underlying Pennsylvanian coal beds. They are in all cases
associated with marine limestones and marls, and the author believes
this fact is of great significance. The gypsum is found in beds ranging
up to roo feet thick and in many places is seen to grade into the limestone.
The deposits in other provinces present no additional features of interest.
The author believes the gypsum comes from conversion of submarine
limestones or marls by the action of free sulphuric acid of juvenile
origin. In support of this theory he points out that numerous circular
blowholes found in massive formations of the gypsum were vents for
escaping gases developed by the action of sulphuric acid on the calcareous
materials.
W. B. W.
Colorado Ferberite and the Wolframite Series. By F. L. Hess and
W. T. SCHALLER. U.S. Geol. Survey,, Bull. 583. Bp. 75;
pls. 14, figs. 35.
In toro the Colorado field, chiefly in Boulder County, furnished
approximately one-sixth of the world’s production of tungsten ore. In
no other field is the iron tungstate the principal ore mineral.
In the first part of the report, Hess discusses the mode of occurrence
of ferberite in this district, the mineral associations being given in con-
siderable detail. He also submits 95 out of 300 analyses examined to
obtain a basis for differentiation from the remainder of the wolframite
group. He proposes the following definition of the group: At one end
of the series shall be placed ferberite, ranging from pure FeWO, to a
composition bearing 20 per cent of the hubnerite molecule MnWO,, and
REVIEWS IOI
at the other end shall be hubnerite in which the proportions of iron and
manganese are the reverse of those given for ferberite. The term wol-
framite shall be reserved for mixtures of these molecules ranging between
the limits assigned to the two end members.
In the latter part of the bulletin Schaller gives a detailed discussion
of the crystallography of ferberite. A total of 32 forms were determined,
12 of which are new for the wolframite group.
W. B. W.
Glacier National Park. By M.R. CAmpBELL. U.S. Geol. Survey,
Bull¥ No:Gco. Pp: 54, figs..3, pls..13-
This bulletin is one of a series intended for popular use, now being
published by the United States Geological Survey. It presupposes no
knowledge of scientific geology on the part of the reader, and is intended
as a guide to the chief physiographic features of the region.
The report takes up a score of the principle valleys, giving a brief
statement for each regarding trails and camps, adjacent mountains,
glaciers, cirques, and other physiographic features of interest. Among
these is the Lewis overthrust fault. It can be observed in most of the
valleys and is a controlling factor in the topography. A thick block of
limestone has been thrust over shales along a fault plane dipping about
10°, for a distance averaging not less than 15 miles. The eastern
boundary of the park follows closely the edge of this overthrust block.
What may be considered the culminating point of the continent is
found on Triple Divide Peak. Waters falling on this peak reach Hudson
Bay, the Gulf of Mexico, and the Pacific Ocean.
Geologists must regret that the scope of this bulletin was not extended
by a few paragraphs on the stratigraphic column exposed in the region.
W. B. W.
Useful Minerals of the United States. By SAMUEL SANFORD and
RALPH STONE. U.S. Geol. Survey, Bull. No. 585. Pp. 250.
Two lists of useful minerals in the United States were published more
than twenty-five years ago in annual reports of the United States Geo-
logical Survey. Many changes in production in recent years require a
new compilation and its publication in more available form.
The plan of the work includes all of the states, and under each is
listed the minerals found and the more important localities. To what
extent the deposits have been mined is indicated in most cases. Data
IO2 REVIEWS
on clays, building stones, and petroleum are included also. The latter
part of the report includes a glossary of more than 4oo terms. Each
definition of a mineral is followed by a list of the states in which it is
found, so that this feature combines the features of glossary and index.
W. B. W.
Geology and Oil Prospects of Northwestern Oregon. By C. W.
WASHBURNE. U.S. Geol. Survey, Bull. No. 590. Pp. 111, pl. 1.
Great development of California oil fields has led to extended pros-
pecting in other regions bordering the Coast Range Mountains.
The sedimentary rocks exposed in this region range from Upper
Eocene to Pleistocene. Shales and coarser clastics of both fresh-water
and marine origin greatly predominate, intercolated with tuffs and vol-
canic agglomerates. Very little detailed work has been done on the
stratigraphy of these systems. Fossils are quite abundant, but there are
few if any remains of diatoms, so abundant in the California oil fields.
The author fails to find indications favorable for oil in this region.
The structure in the northern part is a broad, low geanticline, broken
by many large igneous masses, and by multitudes of small dikes and
faults. That no oil exists is inferred from the fact that in all these
breaks in the strata no true oil seeps have developed. Farther south, in
Coos County and vicinity, the structure is essentially a broad syncline
with low flanking anticlines and few dikes. The structure is favorable
for oil reservoirs, but here also oil-seeps, so abundant in Mexico and
Southern California, are entirely absent.
W. B. W.
Slate in the United States. By T. Netson DALE and OTHERS.
U.S. Geol. Survey, Bull. No. 586. 1914. Pp. 220, figs. 18,
pls. 26.
This report is in the main a corrected and revised edition of Bulletin
275 issued in 1906. Since the publication of that bulletin, slates of
economic value have been found in several states and additional inves-
tigation made in well-known districts.
Part I of the present bulletin summarizes the present knowledge of
the origin, texture, and chemical and mineral composition of slates. The
structure of slate is treated with more detail. In Part II more or less
detailed descriptions are given of occurrences of slate in fourteen different
REVIEWS 103
states, Pennsylvania and Vermont being treated in considerable detail.
Part III takes up the problems of slate prospecting, quarrying, and the
uses of slate.
Statistics for 1913 give the total value of slate production in the
United States as $6,175,476. Pennsylvania produced more than one-
half of the total, and Vermont more than one-fourth.
W. B. W.
Mineral Resources of Alaska. By A. H. Brooxs and OTHERS.
U.S. Geol. Survey, Bull. No. 592. Pp. 413, figs. 13, pls. 17,
map I.
This bulletin is the tenth annual report upon mining conditions and
mineral resources of Alaska. In addition to the administrative report
there are given results of investigations in a score of districts during the
1913 field season. Several of these record the progress made in well-
known mining camps, while others are results of reconnaissance trips in
little-prospected districts. The more important of these preliminary
reports will be embodied in separate bulletins. In these papers emphasis
is laid on conclusions having immediate interest to the miner to whom
a prompt publication is more valuable than a detailed report long
delayed.
Gold continues to be Alaska’s chief source of mineral wealth. The
total production in 1913 was $15,600,000, of which 31 per cent came
from lode mines and the balance from placers. The amount produced
has declined rather steadily since 1906. A marked falling off in 1913 is
attributed in part to unusual scarcity of water during the sluicing season.
The average value recovered from placers has declined from $3.74 per
cubic yard in 1908 to $1.57 in 1913.
Coal is the only mineral product that does not show a decreased out-
put since 1912, and its production is of little consequence commercially.
Tn connection with coal the author states: “‘As a rule, the quality of coal
bears a direct ratio to the amount of deformation, lignite being in least-
folded rocks and anthracite in those most folded.”
W. B. W.
Mineral Production of Canada, for 1913. By JouHN McLEIsuH.
Canada Dept. of Mines. Ottawa, 1914. Pp. 316.
The value of mineral products for the year amounted to more than
$145,000,000, of which over $66,000,000 was in metals. Coal amounted
to over $37,000,000. The remainder is distributed over a large number
104 REVIEWS
of products, none of which approaches in value the amount recorded
for coal.
Especial importance is attached to the quantity of products shipped
from mines and works, the home consumption and the foreign trade.
Joly AD), 13}.
Proceedings of the American Mining Congress. Sixteenth Annual
Session, Phoenix, Arizona, December, 1914. Denver, 1915.
Pp 220:
Contains a detailed stenographic report of the meetings, along with
the text of seventeen papers and addresses presented at the session.
Most of these papers bear on the subject of mining legislation, and
the broader aspects of the economics of the mining and allied industries.
As usual, conservation comes in for its share of discussion.
ACD iB:
The Turquois. By JosrepH E. Pocur. Memoirs of the National
Academy of Sciences, Vol. XII, third memoir. Washington,
LOGS 4. hp 2071S.) 22,7 Mees:
The subtitle reads, ““A Study of Its History, Mineralogy, Geology,
Ethnology, Archaeology, Mythology, Folklore, and Technology.”
The work is admirably adapted to the general reader as well as to the
mineralogist and geologist. A large portion is devoted to the historical
and ethnological study, which is of general interest. Numerous illus-
trations illuminate the text.
The description of all of the known producing localities is of interest
to the geologist and mineralogist.
A. De, B:
Summary Report of Canadian Geological Survey. Sessional Paper
20, to14.) Pp zor maps 3, 1esse
This report contains 40 short papers by members of the staff of the
‘Canadian Geological Survey. Each article is a brief statement of
results of field work in different areas during the 1914 field season. All
of these papers will be supplemented later by more detailed reports.
W. B. W.
A Manual for
Writers
By JOHN M. MANLY
Head of the Department of English of
the University of Chicago
and JOHN A. POWELL
A book designed to aid authors and all others who
are concerned with the writing of English. It aims to
answer the practical questions that constantly arise in
the preparation of manuscripts for publication, busi-
ness letters, and any sort of composition where cor-
rectness of form is an important element.
It treats ina clear and convenient way the matters
of grammar, spelling, and general form about which
writers need to be informed, and gives full directions
on the preparation of ‘‘copy” for the printer and the
correcting of proof. The chapter on letter-writing is
unique and gives just the help that is constantly
wanted and that is not found in other manuals.
viii +226 pages, 12mo, cloth; $1.25 net, postage
extra (weight 1 lb. 8 oz.)
The University of Chicago Press
Chicago, Illinois
Do You Want to Know
What Your Handwriting
Reveals?
ye have published a fascinating
booklet of 32 pages on the sub-
ject of graphology, by William Leslie
French. Every. reader of this publi-
cation will be interested in comparing
the examples of handwriting in judg-
ing character and disposition of the
writer.
A copy of this booklet and 12 differ-
ent patterns of The Spencerian
Steel Pens will be sent postpaid on
receipt of ten cents.
If you can do good work with a poor pen
you can do better work with a good pen.
Silver plated Falcon pens, smooth points,
wear long, will not corrode.
SPENCERIAN PEN CO.
349 Broadway - . - New York City
og PAngyAZZA
==, a
GP
“C0
“Tip 9
—Dupliicge 2.
Sa
ee
=
published recently.
%
December issue of our
354 Fourth Avenue’
\ WHERE DUPLICATE COPIESAREE
IMPORTANT NEW NOVELS
By H. G. Wells, J. C. Snaith, Sir Gilbert Parker, Coningsby Dawson, Booth: Tarkington, William
, J. Locke, F. Hopkinson Smith, and other distinguished American and English writers have been
EDED |
Daus Improved Tip Top Duplicator
is an ideal assistant, always ready to reproduce clean, clear,
perfect copies of letters, manuscripts, price lists, bills, in-
voices, reports, drawings, menus, etc., etc. Don’t take
our word, try it yourself for
Ten days’ trial without deposit
If not satisfactory, return and no questions asked. Makes
150 handwritten and 50 typewritten copies in a few min-
utes. Negatives require‘no washing out after using, and
negative rolls can be used from 15 to 20 times. Complete
outfit contains a continuous roll of Daus Non-Peeling Anti-
Stick Negative Roll. Capsize,8ix13inches. Price $7.5e
FELIX K. DAUS DUPLICATOR CO.
Daus Building, 111 John Street, NEW YORK
Descriptions of these new novels and other important recent publications are contained in the
The Monthly Bulletin of the Latest and Best Books
Sent on request and without charge to anyone interested
THE BAKER & TAYLOR CO.
Wholesale Dealers in the Books of All Publishers
NEW YORK
At Twenty-sixth Street
)
FINE INKS 4» ADHESIVES : BUY i
For those who KNOW ON EXAMINATION?” | —
THAT MEANS) _~ ea Bs:
‘*Test and try before you buy’’ e
ne
“
He.
J e : %
Remington Junior Typewriter vt
Drawing inks : og .
Eternal Writing Ink Remington Quality; Remington Workmanship } —
Bee Engrossing Ink Remington Guaranty ae
. Cee 9 Taurine Mucilage o :
Hi | ] ‘a 8S Photo Mounter Paste Produced purposely to supply the demand for a high- s
Drawing Board Paste grade but low-priced writing machine for the home, ae
Liquid Paste the farm, the store, the office and the study. pe
Office Paste — . .
| Ht esas Vegetable Giue, Etc. Cut out and post today x
} Are the Finest and Best Inks and Adhesives *
REMINGTON TYPEWRITER COMPANY -
(Incorporated) in
327 Broadway, New York
Send me a Remington Junior Typewriter, price $50, a
on free examination. It is understood that I may re- s
turn the machine, if I choose, within ten days. , li I 2
decide to purchase it, I agree to pay for it in 10
monthly payments of $5 each.
Emancipate yourself from the use of corrosive and
ill-smelling inks amd adhesives and adopt the
Higgins Inks amd Adliesives. They will bea
revelation to you, they are so sweet, clean, well
put wp, and withal so efficient.
At Dealers Generally
CHAS. M. HIGGINS & CO., Mfrs.
Branches: Chicago, London
|} 271 Ninth Street Brooklyn, N.Y.
Efficiency, Accuracy
aq Convenience—All at the
© Maximum |
Superiority in general and in detail—built on
the experience of more than 60 years as
optical manufacturers—marks the various models of the
Rausch fomb |
BALOPTICON
THE PERFECT STEREOPTICON Basie eh 3)
In addition to the visible improvements in construction the quality of the optical equipment
and the superior illuminating unit put the Balopticon far ahead of the usual projection lantern-
Model C (illustrated) has the new illuminator—the gas-filled Mazda lamp, surpassing the ©
arc in brilliancy but using less current. Takes all standard lantern slides. Price $37.00.
Models for the projection of opaque objects (photos, postcards, specimens, maps, etc.) $35.00
up. Combined models for both forms, with instant interchange $45.00 up.
Write for illustrated price list containing full description of all models and their uses.
Bausch ¢4 [omb Optical ©.
554 ST. PAUL STREET ROCHESTER, N.%
NEW YORK WASHINGTON CHICAGO SAN FRANCISCO
Leading American Makers of Microscopes, Projection Apparatus, Photographic Lenses, Engineering Instruments,
Stereo-Prism Binoculars, and other High-Grade Optical Products
Me ih
VOLUME XXV NUMBER 2
THE
J OURNAL or GEOLOGY
- A SEMI-QUARTERLY
EDITED By |
THOMAS C, CHAMBERLIN AND ROLLIN D. SALISBURY
With the Active Collaboration of
SAMUEL W. WILLISTON, Vertebrate Paleontology ALBERT JOHANNSEN, Petrology
STUART WELLER, Invertebrate Paleontology ROLLIN T. CHAMBERLIN, Dynamic Geology
ALBERT D, BROKAW, Economic Geology
% ASSOCIATE EDITORS
SIR ARCHIBALD GEIKIE, Great Britain JOSEPH P.IDDINGS, Washington, D.C.
CHARLES BARROIS, France JOHN C. BRANNER, Leland Stanford Junior University
. ALBRECHT PENCK, Germany RICHARD A. F. PENROSE, Jr,, Philadelphia, Pa.
_ HANS REUSCH, Norway WILLIAM B. CLARK, Johns Hopkins University
_ GERARD DEGEER, Sweden WILLIAM H. HOBBS, University of Michigan
' T, W. EDGEWORTH DAVID, Australia FRANK D. ADAMS, McGill University
2 BAILEY WILLIS, Leland Stanford Junior University CHARLES K, LEITH, University of Wisconsin
GROVE K. GILBERT, Washington, D.C. WALLACE W. ATWOOD, Harvard University
; ‘CHARLES D. WALCOTT, Smithsonian Institution WILLIAM H. EMMONS, University of Minnesota
HENRY S. WILLIAMS, Cornell University ARTHUR L. DAY, Carnegie Institution
FEBRUARY-MARCH 1917
ON THE HYPOTHESIS OF ISOSTASY - - - - Se W. D. MacMitran tos
1
ot THE MIDDLE PALEOZOIC STRATIGRAPHY OF THE CENTRAL ROCKY MOUNTAIN
Be: | REGION. I - - - - - - - - - - - C. W. TomtLinson’9 112
se SOME FACTORS AFFECTING THE DEVELOPMENT OF MUD-CRACKS E.M.Kinpre 135
ue DOWNWARPING ALONG JOINT PLANES AT THE CLOSE OF THE NIAGARAN AND
ACADIAN 2 = = > > - - - - - LANCASTER D. BuRLING 145
" THE WESTERN INTERIOR GEOSYNCLINE AND ITS BEARING ON THE ORIGIN
es AND DISTRIBUTION OF THE COAL MEASURES - - Francis M. Van TuyL 150
MA DRCIMAL GROUPING OF THE PLAGIOCLASES - - | = “F.C. Cats” 167
“STUDIES FOR STUDENTS: A CLASSIFICATION OF BRECCIAS - W. H. Norton 160
re ee as
y
bad. UNIVERSITY OF CHICAGO PRESS
CHICAGO, ILLINOIS, U.S.A.
AGENTS
THE CAMBRIDGE UNIVERSITY PRESS Lonpon anp EDINBURGH
KARL W. HIERSEMANN, LErpzic
THE MARUZEN-KABUSHIKI KAISHA, Toxyo, Osaka, Kyoto, Fukuoka, SENDa
THE WISSION BOOK COMPANY, SHANGHAI
U
THE JOURNAL OF GEOLOGa
EDITED BY
THOMAS C. CHAMBERLIN AND ROLLIN D. SALISBURY
With the Active Collaboration of
SAMUEL W. WILLISTON ALBERT JOHANNSEN
Vertebrate Paleontology : 3 Petrology
STUART WELLER ROLLIN T. CHAMBERLIN
Invertebrate Paleontology 3 a Dynamic Geology
ALBERT D. BROKAW
Economic Geology
The Journal of Geology is published by the University of Chicago at the University Press semi-
quarterly, on or about the following dates: February 1, March 15, May 1, June 15, August 1, September 15,.
November I, December 15. ] The subscription price is $4.00 per year; the price of single copies is
65 cents. Orders for service of less than a half-year will be charged at the single-copy rate. | Postage is
prepaid by the publishers on all orders from the United States, Mexico, Cuba, Porto Rico, Panama Canal
Zone, Republic of Panama, Hawaiian Islands, Philippine Islands, Guam, Samoan Islands, Shanghai.
4] Postage is charged extra as follows: For Canada, 30 cents on annual subscriptions (total $4.30), on
single copies, 4 cents (total 69 cents); for all other countries in the Postal Union, 53 cents on annual
subscriptions (total $4.53), on single copies, 11 cents (total 76 cents). {Patrons are requested to make
all remittances payable to the University of Chicago Press in postal or express money orders or bank
drafts.
The following agents have been appointed and are authorized to quote the prices indicated:
For the British Empire: The Cambridge University Press, Fetter Lane, London, E.C. Yearly
subscriptions, including postage, 18s. 8d. each; single copies, including postage, 3s. 2d. each.
For the continent of Europe: Karl W. Hiersemann, K6nigstrasse 29, Leipzig, Germany. Yearly
subscriptions, including postage, M. 22.65 each; single copies, including postage, M. 3.80 each.
For Japan and Korea: The Maruzen-Kabushiki-Kaisha, 11 to 16 Nihonbashi Tori Sanchome,
Tokyo, Japan. Yearly subscriptions, including postage, Yen 9.00 each; single copies, including
postage, Yen 1.50 each. ;
For China: The Mission Book Company, 18 Peking Road, Shanghai. Yearly subscriptions,
9 ee. ee
oo.
2 ee ee ee
$4.00; single copies, 65 cents, or their equivalents in Chinese money. Postage extra, if mailed direct
outside of Shanghai, on yearly subscriptions 53 cents, on single copies 11 cents.
Claims for missing numbers should be made within the month following the regular month of
publication. The publishers expect to supply missing numbers free only when they have been lost in
transit.
Business correspondence should be addressed to The University of Chicago Press, Chicago, Ill.
Communications for the editors and manuscripts should be addressed to the Editors of THE
JOURNAL OF GEOLOGY, the University of Chicago, Chicago, III.
Contributors are requested to write scientific and proper names with particular care and in citations
to follow the form shown in the pages of the Journal.
Separates, if desired, must be ordered in advance of publication. Separates of leading articles will be
printed as ordered, of which 50 (without covers) will be furnished grafzs, the actual cost of the remainder —
(and covers, if desired) to be paid for by the author. The actual cost may vary, and will depend upon
the amount of work in re-making the pages into forms, presswork, paper, binding, etc. Separates
containing half-tones may be expected to cost somewhat more, the increase depending upon the number —
of cuts and the amount of work required upon them.
Entered as second-class matter, March 20, 1893, at the Post-office at Chicago, IIl., under the Act of March 3, 1879
uh
sae
ERRATA
In No. 8, 1916, of this Journal
P. 821, line 11, for Smith read Smyth.
P. 824, line 17, for stopping read stoping.
P. 835, line 10, for Riedemann read Ruedemann.
VOLUME XXV NUMBER 2
THE
JOURNAL OF GEOLOGY
FEBRUARY-MARCH 1977
ON THE HYPOTHESIS OF ISOSTASY
W. D. MacMILLAN
University of Chicago
The splendid papers by Hayford* and jointly by Hayford and
Bowie’ have brought the subject of isostasy into the foreground for
discussion by geologists and others who may be interested. These
papers have taken the subject out of a field of more or less vague
conjecture, and by subjecting it to a very careful quantitative
examination have shown very clearly that isostasy in some form
can be accepted as a reality.
To be sure, they have not proved the reality of isostasy, for in
the mathematical sense no physical hypothesis can be proven.
But they have formulated precise hypotheses of isostasy and have
shown that a vast mass of observational data covering the United
States is very much better satisfied by theories which include their
hypotheses than by the usual gravitational theory which excludes
the hypothesis of isostasy.
Four distinct laws of isostasy have been discussed in these
papers, viz., (a) uniform compensation, (6) uniformly decreasing
tJohn F. Hayford, The Figure of the Earth and Isostasy from Measurements in
the United States, Publications of the U.S. Coast and Geodetic Survey, 1900.
2 John F. Hayford and William Bowie, The Effect of Topography and Isostatic
Compensation upon the Intensity of Gravity, Publications of the U.S. Coast and Geo-
detic Survey, 1912.
105
106 W. D. MacMILLAN
compensation, (c) compensation in a subterranean layer, (d) the
Chamberlin compensation. In so far as merely satisfying the obser-
vations it is found that any one of these hypotheses is as good as
any other, since each of them reduces the sum of the squares of the
residuals by about 90 per cent. This is a very notable reduction
and it places the hypothesis of isostasy upon a solid foundation of
credibility. We say ‘‘credibility” and not “proof” advisedly,
for there still remains the possibility that some other hypothesis,
non-isostatic in its nature, may satisfy the observations even
more closely, and it is a very difficult matter to show that no such
hypotheses exist.
There seems to be nothing inherently improbable in the notion
that the density of the materials under the continents is less than
that under the oceans. MHayford’s success, which must be con-
sidered a notable one, consists in showing, by very complete com-
putations which extend over a large mass of data, that assumptions
of very moderate differences of density are sufficient to bring the
observations and theory into fairly close accord. Whether or not
any other hypothesis will or can be equally successful must of course
be left for the future. Until some such hypothesis makes its appear-
ance we are fairly entitled to put our faith in the broader outlines
of isostasy and leave it to further observations and discussions to
make the details of the theory more precise.
Notwithstanding the fact that all four of the hypotheses dis-
cussed by Hayford satisfy the observations equally well, it would
seem as if Hayford prefers the hypothesis of uniform compensation
to a depth of 122 kilometers and usually has this hypothesis in
mind when thinking of isostasy. This preference, which does not
seem to be warranted by his own discussions,’ is somewhat danger-
ous in that conclusions which are peculiar to this hypothesis are
given a prominence to which they are not entitled. Thus, one is
«Second, it is not possible to ascertain whether this compensation is more
probable than the G compensation, uniformly distributed from the surface to the
depth 70.67 miles, since the two sets of computed deflections agree so closely that
their differences are much smaller than the accidental errors.—The Figure of the
Earth and Isostasy from Measurements in the United States, p. 159.
A corresponding statement is made on p. 162 with respect to the Chamberlin
compensation.
ON THE HYPOTHESIS OF ISOSTASY 107
rather likely to gain the impression from Hayford’s writings that
the “‘depth of compensation”’ is in the neighborhood of 122 kilo-
meters and that this “depth” is as well established as are the
broader outlines of the theory. This is not true, for we do not
know that the compensation is uniform. From the hypothesis of
uniformly decreasing compensation Hayford finds the depth of
compensation to be 175 kilometers, and from the ‘‘ Chamberlin
compensation”? 286 kilometers. Clearly, the ‘‘depth of compen-
sation’’ is very sensitive to change of hypothesis, and it is further
clear that with a slight modification of the hypothesis the ‘‘depth
of compensation”? could be made to retreat to the center of the
earth, or even to vanish altogether." From this it is obvious that
the existence of a precise depth of compensation is not an essential
part of the theory of isostasy. These considerations deprive the
depth of 122 kilometers of the importance or weight which con-
stant repetition is likely to attach toit. It is still doubtful whether
the term ‘‘depth of compensation” corresponds to any physical
reality, however useful the idea may be in our hypotheses.
If the solid portion of the earth were altogether lacking in
rigidity, and if the concentric layers were homogeneous in density,
then the upper surface of the solid earth would be an oblate spheroid,
and this surface would lie about 9,000 feet below the present sea-
level. It would be covered uniformly by the waters of the ocean,
and the pressure at any interior point would be a function of lati-
tude and depth only, and not a function of the longitude. Let us
suppose now that this solid spheroid is endowed with a certain
amount of rigidity and is differentiated somewhat with respect to
density, particularly in the neighborhood of the surface. If the
rigidity were not too great, it seems clear that the heavy regions
would be depressed by the excessive weight, and that the lighter
regions would rise on account of their deficiency of weight. If the
differentiation of density were sufficiently great, it is clear that the
t The idea implied in this definition of the phrase ‘‘depth of compensation” that
the isostatic compensation is complete within some depth much less than the radius
of the earth is not ordinarily expressed in the literature of the subject, but it is an idea
which is difficult to avoid if the subject is studied carefully from any point of view.—
Hayford and Bowie, The Effect of Topography and Isostatic Compensation upon the
Intensity of Gravity, p. to.
108 W. D. MacMILLAN
regions of sufficiently deficient density would eventually appear
above the surface of the waters of the ocean, while the regions of
greatest density would form the great depths of the basins into
which the the waters of the ocean were gathered. Those regions
which have neither excess nor deficiency of density would still
remain about 9,000 feet below the sea-level except in so far as,
through rigidity, they partook of the movement of neighboring
regions. A theory of isostasy comprehending the entire earth in
its grasp should therefore be based upon a level 9,000 feet below
the surface of the ocean. Regions which lie above this level are
deficient in density and regions which lie below this level are exces-
sive in density. Clearly, it would not do to regard the sea-level,
which from an isostatic point of view is an accidental level, as the
dividing surface. Indeed, if the differentiation of density were so
small that the elevated regions were all below the sea-level, we
should be compelled to conclude that the density was everywhere
excessive—a reductio ad absurdum.
But the sea-level is precisely the surface which Hayford has
chosen, and, as a consequence, to all those regions which lie between
the sea-level and 9,000 feet thereunder he has ascribed an excess
of density instead of a deficiency. Since approximately three-
quarters of the earth’s surface lies under the ocean at an average
depth of approximately 15,000 feet, and since under Hayford’s
hypotheses too great density is ascribed to this region, it would
seem that his hypothesis has had the effect of raising the mean
surface density for the entire earth from 2.67 to a somewhat higher
figure. This alone might not be of any great importance, since
at best the figure 2.67 is somewhat uncertain, but a hypothesis
which is based upon sea-level gives horizontal changes of defect of
density over the continents which are relatively too great, as is
shown in Table I, which is based upon the hypothesis of uniform
compensation.
Thus, with the sea-level basis the defect in density under a
region which has an altitude of 6,000 feet is six times as great as
the defect in density under a region which has an elevation of 1,000
feet, while with a true, isostatic theory the defect would be only
one and one-half times as much. The defect under a 1,000-foot
ON THE HYPOTHESIS OF ISOSTASY 109
elevation is twice as great as under a 500-foot elevation on the one
hypothesis and only twenty-nineteenths times as great on the other.
This indicates that the sea-level basis overaccentuates the impor-
tance of changes of level in the topography of the continents.
TABLE I
eee = Defect in Densi-|Defect in Densi- nes Ye Defect in Densi-|Defect in Densi-
Elevation in | ty Sea-Level | ty 9,000-Foot Hlevation | ty Sea-Level | ty 9,000-Foot
co Basis Basis ° Basis Basis
Ome te, fo) 9 ALCOOs i.e 4 13
THOOOW ea eae it ite) GROOOnmeraman 5 14
BEOOO Saas | 2 TL Ons ase 6 15
EROOOH anes sia: 3 12 WNOOOEsin este 7 16
If the density in the earth’s crust actually varies in the manner
supposed by Hayford (using the sea-level basis), one would expect
regions over which the compensation was effected to be smaller
than if the density varies according to a true isostasy (i.e., the
g,000-feet-below-the-sea basis), since the changes in density are
relatively greater in the first case than in the second. Hayford
has attempted to determine the sizes of these areas of compensation,
. but the quantities to be considered were so small that success was
scarcely to be expected; indeed, they are ‘‘frequently less than the
errors of observation and computation” and, possibly, also less
than the effects of local irregularities of density. The evidence,
though inconclusive, leaned faintly toward rather small areas of
compensation, and Hayford is of the opinion that these areas are
between a square mile and a square degree.’ If the present writer
is correct in assuming that a true isostasy must be based upon a
level 9,000 feet below the sea, then the evidence published by
t It is certain from the results of this investigation that the continent as a whole
is closely compensated, and that areas as large as states are also closely compensated.
It is the writer’s belief that each area as large as one square degree is generally largely
compensated. The writer predicts that future investigations will show that the
maximum horizontal extent which a topographic feature may have and still escape
compensation is between one square mile and one square degree. This prediction
is based, in part, upon a consideration of the mechanics of the problem.—Hayford
and Bowie, The Effects of Topography and Isostatic Compensation upon the Intensity
of Gravity, p. 102.
I1IO W. D. MacMILLAN
Hayford and Bowie on the extremely delicate question’ of the
“areas of compensation”’ is not altogether trustworthy, and the
sizes of these areas must be regarded as unknown.
From the fact that the hypothesis of isostasy reduced the sum
of the residuals from 65,434 to 8,013, or by approximately 90 per
cent, and from the fact that the average elevation of the United
States is about 2,500 feet Hayford concluded that the average
departure from complete isostasy in the United States is equal to
about 250 feet of rocks. It is not easy to see how Hayford drew
this conclusion. It certainly has no mathematical justification,
for even if the theory were perfect and the isostasy complete, the
sum of the squares of the residuals would not be zero, since the
imperfections of the observations would still give us a very respect-
able, but quite unknown, sum. How then can we form a quanti-
tIf the separate anomalies in the United States be compared, it is found that
in 16 cases out of 41 the anomaly with local compensation assumed is smaller than
with regional compensation assumed uniformly distributed to zone K (18.8 kilo-
meters), and only 13 cases in which it is larger. Similarly, there are 20 cases out of
41 in which the anomaly with local compensation is smaller than with regional com-
pensation extending to zone M (58.8 kilometers), and only 15 cases in which it is
larger. There are 26 cases out of 41 in which the anomaly with local compensation
assumed is smaller than with regional compensation assumed to extend to zone
O (166.7 kilometers), and only 12 cases in which it is larger. In all other cases the
two anomalies compared are identical to the last decimal place used, the third.
The evidence either for or against local compensation in comparison with such
regional compensation distributed uniformly over these moderate distances is neces-
sarily slight and possibly inconclusive. For, as shown in the table, the difference
between computed effects of compensation in the two cases compared is very small
upon an average. The whole evidence is furnished by these very small differences,
which frequently are less than the errors of observation and computation. As shown
by the table, there is but one station among the 41—namely, No. 43, Pike’s Peak—
at which the difference between the computed effect of local compensation and the
computed effect of regional compensation uniformly distributed to zone K exceeds
0.004. Such a difference tends to become greater as the distance over which the
regional compensation is supposed to be uniformly distributed is increased, but
columns 7 and 8 of the table show that even when the regional compensation is assumed
to extend to zone O,a distance of 166.7 kilometers, from the station, there is only one
station among the 41—namely, station no. 54, San Francisco—at which the computed
effect of local compensation and the computed effect of regional compensation exceeds _
©.017 dyne.
Nevertheless the evidence, slight as it necessarily is, indicates that the assump-
tion of local compensation is nearer the truth than the assumption of regional com-
pensation uniformly distributed to zone K (18.8 kilometers). The evidence is still
stronger in the same direction when the comparison is made between local compensa-
tion and regional compensation extending uniformly to the greater distances, 58.8
and 166.7 kilometers, represented by zones M and O.—Hayford and Bowie, The
Effects of Topography and Isostatic Compensation upon the Intensity of Gravity, p. 101.
ON THEVLVPOTHESTS OF TSOSTASY TEALA
tative judgment from the sum of the squares of the residuals which
depends, not only upon the imperfections of the theory, but upon
the imperfections of the observations as well? Obviously, it
cannot be done. The estimate of 250 feet is little more than a
guess, however shrewd the guess may be. If we use the 9,000-foot
level as the basis of our guess, then the average elevation of the
United States is 11,500 feet, and the average departure from com-
plete isostasy is 1,150 feet of rocks instead of 250 feet. It would
not be altogether fair, however, to make this direct substitution,
for the reduction of the sum of the squares of the residuals from
65,434 to 8,013 was accomplished on the sea-level hypothesis, and
the reduction might be quite different under another hypothesis.
From a purely mathematical point of view, any set of a finite
number of observations of the intensity and direction of gravity
can be satisfied, not approximately, but exactly, in infinitely many
ways by a proper distribution of density in the earth. The virtue
of the theory of isostasy, therefore, lies, not in the mere fact
that the observations are more nearly satisfied by the theory
than without it, but in the fact that a definite principle is laid
down for the variations of density, and that this principle brings
theory and observations into a satisfactory accord. As Hayford’s
four distinct hypotheses show, any smoothly uniform hypothesis
of isostasy can be regarded only as a first approximation to the
actual situation, and Hayford has been successful in showing that
any one of these four hypotheses is a good first approximation. It
is equally clear that such delicate points as “depth of compen-
sation’”’ and size of “areas of compensation”’ depend for their suc-
cessful determination upon the vastly more difficult matter of
second and higher approximations, and these approximations can
be obtained, if at all, only by a very much more dense net of obser-
vations, and quite likely the observations themselves would have
to be still further refined. ;
While the theory of isostasy has made a very successful approach
to the solution of the problem of bringing the anomalies of obser-
vation into accord with the theory of gravity, it must be admitted
that there is no evidence to show that the solution of the problem
is necessarily isostatic.
THE MIDDLE PALEOZOIC STRATIGRAPHY OF THE
CENTRAL ROCKY MOUNTAIN REGION:
C. W. TOMLINSON
University of Chicago
CONTENTS
PART I
FOREWORD
Locations of sections
STRATIGRAPHY
The Correlation Diagrams
Use of the Diagrams
Method of Correlation
Source of Data
Standard List of Members
Mississippian
Devonian
Silurian
Middle and Upper Ordovician
Upper Cambrian and Early Ordovician
Paleontological Collections: Assembled Lists
Disconformities
PART II
STRATIGRAPHY—C ontinued
Upper Cambrian and Early Ordovician
Members 0 and 1
Members 3 and 4 in Northern Utah
Correlation with the Pogonip Group of Nevada
Member 3 in Wyoming and Montana
The Maxfield Formation of the Central Wasatch
The Close Relation of the Canadian Series to the Upper Cambrian
Series
Is the Ozarkian System Represented Here ?
The Ordovician Quartzites and Sandstones
The Eureka and Swan Peak Quartzites
The “Ogden Quartzite” in the Uinta Range
The Sandstone at the Base of the Bighorn Formation
t An abridgment of a thesis presented for the degree of Doctor of Philosophy at
the University of Chicago, 1916.
112
PALEOZOIC STRATIGRAPHY OF ROCKY MOUNTAINS 113
The Middle and Upper Ordovician Dolomites
Extent of the Bighorn Dolomite
The Bighorn Dolomite in Montana
Members 2-4
Members 5-7; the Leigh Formation
Members 8 and 9
PART III
STRATIGRAPHY—Continued
The Silurian System
In Utah
In Nevada
Is the Laketown Dolomite in Part Devonian ?
Pre-Laketown and Pre-Devonian Emergence
The Devonian System: The Jefferson Dolomite
Basal Division of the Jefferson Dolomite
Main Division of the Jefferson Dolomite
Upper Divifion of the Jefferson Dolomite
Misuse of the Name “Jefferson” in Yellowstone Park and Vicinity
Correlation of the Jefferson Dolomite with the Nevada Limestone
The Devonian System: The Three Forks Formation
Type Sections
Members 19 and 20
Beds of Doubtful Age in the Bighorn Range and near Cody
‘Member 21
The Three Forks Formation in Northern Utah; Correlation with the
Benson and Ouray Limestones
Disconformities
Summary of Results on Special Problems of Investigation
HISTORICAL SKETCH
Upper Cambrian and Lower Ordovician
Extent of Submergence
Paucity of Clastic Sediments
Evidence That the Sea Was Shallow
The Chazyan(?) Sands
Emergence and Peneplanation
Middle and Upper Ordovician
The Trenton Submergence
The Post-Trenton Emergence
The Richmond Submergence
Silurian
Emergence
The Niagaran( ?) Invasion
II4 C. W. TOMLINSON
Devonian
Pre-Jefferson Erosion
Devonian Marine Invasions
The Upper Devonian Muds
The Pre-Mississippian Interval of Emergence
Depth of Erosion
“Positive” and ‘‘ Negative” Areas: Axes of Warping
PART I
FOREWORD
This paper is the result of three months’ field work by the writer
in the central Rocky Mountain region during the summer of 1915,
under the auspices of the University of Chicago. For assistance in
planning the field investigation the writer is indebted to Dr. Eliot
Blackwelder, of the University of Wisconsin, and to Dr. R. D.
Salisbury, of the University of Chicago; for aid in the identifica-
tion of fossils, to Dr. Stuart Weller, of the University of Chicago;
and for helpful suggestions and criticism of the material of the thesis,
to all of these gentlemen.
The problem centered about several broad gaps in the existing
knowledge of the Ordovician, Silurian, and Devonian history of
western Wyoming and adjacent parts of neighboring states; for
example: (1) the ‘‘Jefferson limestone” of the Absaroka Range,
of Yellowstone National Park, and of southwestern Montana was
known to be in part Devonian, but the presence of Ordovician and
Silurian strata within this formation and its relation to the Bighorn
dolomite of central and north-central Wyoming were matters of
dispute; (2) hiatuses were suspected at the base and at the top of
the Ordovician-to-Devonian sequence of this region, and at more
than one horizon within that sequence, but no physical evidence of
any such hiatus ever had been cited; (3) the relation of the Wyo-
ming Ordovician and Upper Cambrian to the corresponding systems
in northeastern Utah had never been studied by careful strati-
graphic comparison; and (4) the relation of the Silurian to the
Ordovician in northeastern Utah was unknown.
By first-hand studies, in one season, of ten complete, and in
most cases excellently exposed, sections at strategic localities
scattered throughout the area involved, the writer is enabled to
throw considerable new light upon all of these questions.
PALEOZOIC STRATIGRAPHY OF ROCKY MOUNTAINS 115
Because of the well-known scarcity of fossils in the strata which
were to form the subject of his investigation, the writer set out with
intent to make the greatest possible use of correlation by means
of lithologic characters. To this end he made accurate measure-
‘ments and described in detail, in every section, each member of
the sequence which could be distinguished from other members by
its lithological characters. Collections of fossils, although made
secondary to the work of lithological description and measurement,
and in no case exhaustive, were made wherever opportunity pre-
sented itself, and in each case served to corroborate the correlation
which otherwise would have been made on the basis of lithology
alone. For instance, the presence of the Ordovician Bighorn
dolomite as the thick basal member of the so-called “ Jefferson
limestone” in the areas of the Absaroka, Wyoming (No. 52), and
Livingston, Montana (No. 1), folios was established beyond doubt
by both lines of evidence, though it would have been well established
by either one alone.
In these correlations it was realized that the coincidence of the
lithological characters of a single member at one locality with those
of one member in the same stratigraphic position in another locality
is much more inconclusive (although significant) than the corre-
spondence of several successive members in one section to the same
number of members occurring in the same order and in the same
stratigraphic position in another. In nearly every case where a
single bed in one section was found to correspond accurately in
character to one bed in another section, not less than two other
contiguous members were found to correspond in like manner.
Where several members of one section appeared to be missing from
another section, a hiatus was inferred in the latter; and such
inference was substantiated in several cases by more direct evidence
at the suspected horizon.
LOCATIONS OF SECTIONS MEASURED, 10915
(See map, Fig. 1)
I. (Partial.) In the northwest wall of the canyon of Big Goose Creek,
about 20 miles southwest of Sheridan, Wyoming.
II. Goose Creek Ridge-—On the crest of the ‘“‘limestone front ridge” on
the northeast flank of the Bighorn Range, between Big and Little Goose
creeks, about 20 miles southwest of Sheridan, Wyoming.
C. W. TOMLINSON
116
III. Cody, Rattlesnake Mountain.—On the southwest angle of Rattlesnake
Mountain, north of Shoshone Dam, 9 miles west of Cody, Wyoming.
IV. Dead Indian Creek.—On the nose of the main ridge running parallel
to the valley of Dead Indian Creek on the southeast side thereof, just south-
godvd sty} Ul 0} polIayjor sorpyeoo] Surmoys deyy—'t “ory
*swvsbvIp u0172 191109 | ]
e ; aud Curqonijsuos ui pasn ; aug
Nouv j SsoyyOo fq P2eqsrsap $u0rj,96 O|! WUAe
ee “SIG] UI UosSuwoy md |
Ra pasnsvaur s4011935 ©
v
‘aN3931 q62n3/2
O.dv ad O 7 O J thane uokuvy
eee ‘a Gig yL04'5 @ wd.
Bho s f Lon © el
|
ena Hemswodig
aiuoy 2OMY PUM PVan
Sh uEsR |
i
j uote) JOA PUM res ee
» © ze
i aoayy spac foie WD VROP at Retna
9 SAVY APA ‘ npoy
ussyenes ‘ UG
5 sywusaiz)0y
SH OCIS * 31 ;
wl ie eet tN seoesgeaes yaasg"""@ NS a:
ab pry 4224) 25099 © Uorpul puadea, é! bag vatauy~,
|
|
|
On 7G
Playoag Raaang
ureyUNnoW KRmoug
ywey Uojsburrry 2...
Vo Ne Vv. NSO WW
“S4W 212 FIVNT¢
west of the crossing of the road from Cody to Sunlight Basin, Wyoming, a
few miles east of the east border of the Crandall quadrangle.
V. Crandall Creek.—At the west end of the north slope of Windy Moun-
tain, in the south wall of the valley of Clark Fork, near the mouth of Crandall
Creek, in the Crandall quadrangle (Absaroka Folio), Wyoming.
PALEOZOIC STRATIGRAPHY OF ROCKY MOUNTAINS 117
VI. (Incomplete.) Antler Peak, Gallatin Range, Yellowstone Park,
Wyoming.
VII. (Cambrian only.) On the ridge running west from the summit of
Livingston Peak, Livingston quadrangle, Montana, about one mile north of
“Old Baldy.”
VIII. Livingston Peak.—On the south slope of Livingston Peak, on the
ridge leading to ‘‘Old Baldy” (the 9,500-foot peak 1 mile southwest of
Livingston Peak), Livingston quadrangle, Montana.
IX. Logan, Montana.—On the ridges north of the Gallatin River, opposite
Logan, in the Three Forks quadrangle, Montana.
X. Teton River.—On the divide between Teton River and South Leigh
Creek, Grand Teton quadrangle, Wyoming.
XI. Blacksmith Fork (including the Cambrian).—Measured across the
ridges just north of the canyon of Blacksmith Fork, Cache County, Utah, from
Cottonwood Gulch to the crest of Logan Peak, west of Saddle Creek.
XII. Manitou.—Measured on the ridges from one to two miles northwest
of Manitou, Colorado.
STRATIGRAPHY
THE CORRELATION DIAGRAMS
Use of the diagrams.—Detailed descriptions of the 12 sections
above listed cannot be printed here. A consistent application of
the principles of correlation outlined on pp. 115, 122, however, has
made it possible to formulate a standard list of members. Each
of the 12 sections is made up of a part or all of the members in this
list (see pp. 123-28), and includes no others. By combining these
members in the manner indicated by the accompanying correla-
tion tables (Figs. 2, 3) and diagrams (Figs. 4, 5), every one of the
17 stratigraphic sections used in the compilation of the list can be
reconstructed. For instance, the Devonian system in the Crandall
Creek section is made up of Member 1, with a thickness of 47 feet,
overlain by Member 2, 26 feet thick, which is followed by Member 3,
28 feet thick, and so forth. Even if some of the strata in a given
section be erroneously placed in the correlation table, yet the table
and the standard list of members will furnish a correct description
of that section, with accurate measurements of its constituent
parts.
The description of each member is given in considerable detail,
but is sufficiently generalized in each case to cover all observed
variations in character from place to place. In the few cases where
TOMLINSON
W.
C.
ioe)
oi
BaEWn3aW
sula}SAs UBINIS pue “URITAOPIO ‘URIIqUIeD JY} 1OJ B[Gv} WOTV[IIIOQ—'% “O1T
“SSaUyr1yy ayPr2?pUr saanhr} sabavy ayy
a
°SM001I9S paqgrysdIsap ayy Uy Spaq opsajaa Soseyjuaund u? saunhry
slel
| 03" |
*,09 | pain
9) -svow
Ose
F,0vl e
(6-2) (-L) FZ)
OO
| rue
24 9 a
D
XXX) DAN
(Aao>)
NIV |33Y¥9] W33N9) Nive | 4V3d
aN NVIGNI] WIva
35009} _sariyy] GYIG|-NvUd| AMONS|-ONIAIT
6Ee
(u-2)
|
3sou
-13W
Uosur}ULO} "AA'D RQ pains paw u01j295
asodxg JON
Of,
FX
i
xX
(4299) 1334
maj ‘LD204
Zs
Raa
peel
g 3)
£6
(e~£e a3)
AS
6e-5<-a'3
Fa
Lg z
><
28-04,
OLl
=00S
¥,00S
DPIWOLCT be a.
UMoayDD UdsArY ys24 0
spwuojog
Ge68|
ILE
96-%6
vz0l-66
O79
Sa,PI1pur
[o>] 0] >| of of ef ro] |
fswon| = 20
feersina] ada
y¥3aMo1
NYOHDIGA
NOLLWWYOJ INMOLINVI
A 09 a@uH4 oO
2)
N V
119
PALEOZOIC STRATIGRAPHY OF ROCKY MOUNTAINS
|
h |
H | wbt
f (it-98¢)
1 i}
Secey
NiNeSES
AV Ni cSt Lit?
(19-95)
¥,0S5
3u04s
Meas, 4| F000!
dosip | (e2ea)
He
*
pg
|
if |
Y |
S|
9 |
2 |
8 |
6 |
Or
uy
RT
et)
+
Eu
Eu
2
EX
ey
oe
re
ra
UI9}SAS ULTUOAD(T IY} IOJ 9IQe} UOTJe[AIIOQ—'E “OIy
tote
F00S | ,008
(69) (a)
eX
0OSIF |. fF 00S!
(oy _|-008t |-ooxt]} ” ~00%
persons @rSon) T (224) | ©) Tova
\/
X
/\
00
(ie +31
L
Sel
(1-9)
es
X
,00£L
()
R (gee
en en
2.8 ‘ay Ueaeaey
jorerg) Jo 22d
a pW0j0d uosuaffop
,voe
(vz- 119)
Fae
£va)
oS
Gea
091
2Y-S¥a
+,0071
92}
(Ov-Le) |(vé-0£ 9)
wi b-T,
>< Gt)
Xx
-0S2 | oo21
(eb-sea)|-009
200%
eer
ty
+8L
(Gri~ Si)
eum
(vs1-05))|
pay wy:
UNO 2uoy
fo yard
BuUuo,sswry VWDPvVASGB|
¥,0Q0S
é
Nig OMeSnuree sasha i
Ered Ga SEE
coisa |NViddISSISSINee-»
120 C. W. TOMLINSON
ANDOLPH
[ase Tvincs-
TOR PLAK
RICHMOND
EUREKA - SWAN PEAK
LEGEND FOR THIS AND THE ACCOMPANYING DIAGRAMS.
Boundary between UThologic Types clearly Y Stratigraphic position uncertain because of
¥ i defined. ; lack of defail in published descriptions.
te Disconformity between strata above and be- NS Lithologic characters typical; position uncer-
low this horizon established on physical BS Tain because of isolation inthe sequence
evidence. : and lack of fossils.
Does Disconformuy between strata above and below Litholagic characters not fully Typica) of this
this horizon suggested by physical evidence. member; stratigraphic position therefore
Strata present : in doubt:
e Member or members between nearest defined a0 No substantial evidence of hiatus between
boundaries are represented by strata with these members noted in any of these sections
thickness tndicaccd b het ght of column. Reduced thickness of members net improbably
= Strata thus shown may include répresentatives of due to slower rate of deposition. Inthe
5 higher or lower jembers (according to the Ordovician, Yntmber 9, where shown, may be con-
position of the dot) as wellas of those direct! temporaneous with part of member S of other
indicated ; but no vepreseritatives of members seetions, Inthe Devonian of Ceo
beyond the next defined boundary are inclu- and Dead Indien Creeks, member I may b
ded. Uncertainty due to lack of detart in pub- contemporaneous with part ef member 2 of
lished descviptions, or To poor exposure. the Blacksmith Fork section. P
White spaces : Strata missing frem sachlon The background is stippled To throw the sections trite relief.
Fic. 4.—Correlation diagram for the Oidovician system, drawn to scale
I21
PALEOZOIC STRATIGRAPHY OF ROCKY MOUNTAINS
gos 0} UMVAP ‘t9}SAS URIUOAI( IY} OJ WeAseIP UOTVPIaIIOJ—S OL]
Bf See 3 5 1
KG
2 RH BRR: Bae eee
ss
No0oS uw 3A ABE
122 C. W. TOMLINSON
such variations are conspicuous or important they are specifically
described for the locality where they were noted.
The standard list of members, together with the detailed correla-
tion tables (Figs. 2, 3), in effect furnishes a compound section for
17 different localities in the central Rocky Mountain region.
Method of correlation —In making the correlations-here set forth
the notes and the specimens from each section were carefully com-
pared with the notes and the specimens from every other section,
at first singly, then in groups. When a thorough and comprehen-
sive tentative plan of correlation had been completed, all of the
specimens from all the sections were laid out on a specially prepared
floor, marked off into intersecting columns and rows, so that com-
parison of all the specimens thought to belong to a given horizon
could be made at once, and at the same time all possible alterations
in the proposed plan of correlation could be considered with the
specimens in view. Happily, the original plan stood the test with
very few changes, and those were of minor consequence.
All available paleontological data were carefully studied to
determine as accurately as possible the age of the several members.
This correlation is not to be considered as final in all details, and
even a few of its larger features are still clouded by uncertainty
due to lack of data. The points in doubt are indicated in the dia-
grams and discussed in the pages following.
Sources of data.—The 7 sections, other than those measured by
the writer, which are embodied in the correlation were taken from
the following sources:
Eureka District, Nevada.—Arnold Hague, ‘Geology of the Eureka District,
Nevada,” U.S. Geol. Survey, Monographs, XX (1892).
Randolph quadrangle, Utah.—G. B. Richardson, ‘‘The Paleozoic Section
in Northern Utah,” Amer. Jour. Sci., 4th Ser., XXXVI (1913), 406-18.
Labdarge Mountain, Wyoming.—Eliot Blackwelder, unpublished manu-
scripts, U.S. Geol. Survey; E. M. Kindle, “The Fauna and Stratigraphy of the
Jefferson Limestone in the Northern Rocky Mountain Region,” Bull. Amer.
Pal., IV, No. 20 (1908), 12-13.
Survey Peak, Wyoming.—J. P. Iddings and W. H. Weed, “Descriptive
Geology of the Northern End of the Teton Range,” U.S. Geol. Survey, Mono-
graphs, XXXII, Part 2 (1899), chap. iv, p. 180.
Princeton, Montana.—E. M. Kindle, op. cit., p. 10.
Melrose, Montana.—E. M. Kindle, op. cit., p. 9.
PALEOZOIC STRATIGRAPHY OF ROCKY MOUNTAINS 123
Snowy Mountain, Wyoming.—W. H. Weed, “Geology of the Southern
End of the Snowy Range,” U.S. Geol. Survey, Monographs, XXXII, Part 2
(1899), chap. vi, pp. 206, 213-14.
With the exception of Blackwelder’s section at Labarge Moun-
tain none of these was described in detail, and the correlation given
for them is of necessity very general and subject to thoroughgoing
revision.
Additional data on the thickness of Cambrian and Mississippian
members were secured from the following sources:
Teton River, Wyoming.—Eliot Blackwelder, unpublished manuscripts,
U.S. Geol. Survey.
Antler Peak, Wyoming.—J. P. Iddings and W. H. Weed, ‘Descriptive
Geology of the Gallatin Mountains,” U.S. Geol. Survey, Monographs, XXXII,
Part 2 (1899), chap. i, p. 22.
Logan, Montana.—A. C. Peale, ‘‘ Description of the Three Forks (Montana)
Sheet,”’ Geol. Atlas U.S., Folio 24 (18096).
Livingston Peak, Montana.—Arnold Hague, “‘ Description of the Livingston
(Montana) Sheet,’’ Geol. Atlas U .S., Folio 1 (1894).
- Crandall Creek, Wyoming.—Arnold Hague, ‘Description of the Absaroka
Quadrangle,” Geol. Atlas U.S., Folio 52 (1899).
Rattlesnake Mountain, Wyoming.—C. A. Fisher, “Geology and Water
Resources of the Bighorn Basin, Wyoming,” U.S. Geol. Survey, Prof. Paper
No. 53 (1906), p. 14.
Goose Creek Ridge, Wyoming.—N. H. Darton, ‘Description of the Bald
Mountain and Dayton Quadrangles,” Geol. Atlas U.S., Folio 141 (1906).
Princeton, Montana.—W. H. Emnions and F. C. Calkins, “Geology and
Ore Deposits of the Philipsburg Quadrangle, Montana,” U.S. Geol. Survey,
Prof. Paper No. 78 (1913).
Data concerning the Three Forks formation near Logan,
Montana, were taken from: | :
W. P. Haynes, “The Fauna of the Upper Devonian in Montana, Part 2,
The Stratigraphy and the Brachiopoda,” Annals of the Carnegie Museum, X
(1916), 16-17.
STANDARD LIST OF MEMBERS
CONSTITUTING THE MIDDLE PALEOZOIC SECTION IN WESTERN WYOMING,
SOUTHWESTERN MONTANA, AND NORTHEASTERN UTAH
(M.T.= Maximum thickness known in this region)
MISSISSIPPIAN
3. Main body of the Madison limestone. Interbedded massive and flaggy
limestones, blue-gray, gray, brown, or black, crystalline to dense, with some
layers abundantly fossiliferous. Chert of local occurrence only. M.T.,
124 C. W. TOMLINSON
2,000 feet, more or less, in Utah. 1,500 feet in the Livingston quad-
rangle, Montana.
2. Coarsely crystalline, very fossiliferous limestone, free from chert. M.T.,
150 feet, Snowy Mountain section, Yellowstone Park.
rt. Very hard, cherty, black or gray limestone. Fossils few or fragmentary.
M.T., 175 feet, Snowy Mountain. (Members 1 and 2 of the Mississippian
are differentiated from Member 3 only in Yellowstone Park and vicinity.)
DEVONIAN
(Member 21 may prove to be of Mississippian age; Members 1 and 2 may
be Silurian.)
Three Forks Formation
(M.T., 978 feet, Blacksmith Fork)
21 (=Haynes’s 1, 2, and 3). Thin-bedded or papery black shales, overlain
by yellow or reddish sandstones, sandy shales, or arenaceous limestones.
M.T., 197 feet, Blacksmith Fork.
BH ens See oislisy Tee Break in Sedimentation RAPPER ee
20B. Blue-gray platy or nodular limestone. M.T., 361 feet, Blacksmith
Fork (Beds 146B-149).
20. (=Haynes’s 4 and 5). Fissile green shale, capped by nodular gray lime-
stone, both very fossiliferous; the horizon of the Clymenia fauna in
Montana. 130 feet, Logan, Montana. M.T., of Members 19 and 20
together, 420 feet, Blacksmith Fork (Beds 146-146A).
19. (=Haynes’s 6 and 7). Orange-yellow to reddish platy limestones and cal-
careous shales (drab to cream or yellow on fresh surfaces). 109 feet,
Logan (238 feet, Goose Creek Ridge).
Upper Division of the Jefferson Dolomite
(M.T., 270 feet, Blacksmith Fork)
18. Brecciated (4-inch to ?-inch fragments) drab to gray-brown, massive
dolomite, in most places forming cliffs. M.T., 136 feet, Blacksmith Fork.
17. Drab, yellow, and buff to dark-brown, thin-bedded limestones or dolomites,
locally in part shaly. M.T., 85 feet, Livingston Peak.
16. Moderately massive dolomite, gray to brown, with rough surface. Spar-
ingly fossiliferous in Yellowstone Park. M.T., 40 feet, Antler Peak
(Iddings and Weed).
15. White or variegated sandstone, locally represented by light-gray dolomite
full of coarse quartz grains, or by limestone conglomerate(?) (Snowy
Mountain). In the Teton River section, interbedded with dark-brown,
PALEOZOIC STRATIGRAPHY OF ROCKY MOUNTAINS 125
pitted dolomite, probably representing part of Member 16. 10 feet,
Labarge Mountain; 51 feet total, Teton River.
Break in Sedimentation
Main Division of the Jefferson Dolomite
(M.T., 681 feet, Labarge Mountain)
14. Imperfectly exposed. Ledges of fine-grained, brown dolomite. Float
of thin-bedded, drab and buff dolomite also. M.T., 160 feet, Labarge
Mountain.
13. Cliff-making, steel-gray or gray-brown, fine-grained, calcitic, pitted dolo-
mite. M.T., 65 feet, Blacksmith Fork.
12. Very poorly exposed. Float of thin-bedded, light-gray and buff dolomite,
very calcitic, grayish-brown dolomite, and calcareous shale. M.T., 72
feet, Labarge Mountain.
8-11. Main body of the Jefferson dolomite in Wyoming and Montana.
Blackish to light-brown and brownish-gray, thick-bedded dolomite,
weathering to brown-gray or gray. Very fetid odor on fresh fracture.
Characteristically with a rough weathered surface. Locally calcitic, or
with some layers mottled and streaked with buff. Locally quite fossilifer-
ous at certain horizons. 312 feet, Logan; probably somewhat more at
Labarge Mountain and at Blacksmith Fork.
10. (Differentiated at Antler Peak only.) Brown and drab to whitish dolomite
with yellow-brown bands. M.T., 75 feet, Antler Peak.
g. Dark-brown, saccharoidal, fetid dolomite, with rough weathered surface.
M.T., 28 feet, Antler Peak.
8B. White, drab, or pearl-gray dolomite, conspicuous because of its light color.
M.T., Labarge Mountain (Bed 16), 3 feet; Teton River (Bed 26), 1 foot
6 inches; Antler Peak (Bed 10), 3 feet; Crandall Creek (Bed 43), 2 feet
6 inches.
8A. Light-brown to blackish-brown, fine-grained dolomite. M.T., 28 feet,
Antler Peak.
6-7. Dense to finely crystalline, white, cream, or pale-gray dolomite, platy
to blocky. M.T., 37 feet, Teton River. In the Crandall Creek section
sedimentation appears to have recommenced with Member 7 after an
interval of nondeposition.
5. Fine-grained or saccharoidal, dark-brown, fetid dolomite, with irregular,
pitted, weathered surfaces. M.T., 23 feet, Dead Indian Creek.
4. White friable sandstone at Labarge Mountain, elsewhere represented, like
Member 15, by a bed of quartz grains more or less closely packed together
in a matrix of white or yellowish dolomite. M.T., 10 feet, Labarge
Mountain.
ORS NA ORES Nacht Break in Sedimentation (?) 2
126 C. W. TOMLINSON
bS
Basal Division of the Jefferson Dolomite
(M.T. for Wyoming, tor feet, Crandall Creek)
Fetid, dark-brown, saccharoidal dolomite, mostly thin-bedded. M.T.,
95 feet, Labarge Mountain.
. White, light-drab, or very pale lavender, dense or finely crystalline dolomite,
breaking up into small, sharply angular talus. Locally carries ostracods..
M.T., 202 feet, Blacksmith Fork. (Probably included with the Silurian
by Kindle and Richardson in Utah because of its light color. Possibly of
Silurian age.) Lies directly on the Ordovician in many sections, locally
with marked disconformity.
. (Typically developed in the Crandall Creek section only.) Thin-bedded,
greatly variegated dolomite, with thin bands of red shale and black chert.
Dolomites saccharoidal to very dense and closely laminated. Thin lime-
stone conglomerate at the base. Lies directly upon the Ordovician at
Crandall Creek. M.T., 47 feet, Crandall Creek.
; . Break in Sedimentation . :
(In Wyoming and Montana; not in Utah ?)
SILURIAN
(Not known in Wyoming, Montana, or Colorado; typically developed at
Blacksmith Fork, Utah, where it includes the following members:)
8.
Massive, brown-gray dolomite, medium finely crystalline, with rough
weathered surface. 190 feet.
. Thin-bedded, slabby limestone. 5 feet.
6. Massive, pearl-gray or whitish limestone, medium finely crystalline, with
hackly talus; lower 45 feet contains abundant nodules of snow-white chert
like Member 5. 140 feet.
. Massive white chert, with many quartz geodes. to feet.
. Extremely massive, cream-colored, coarsely crystalline dolomite. Upper
20 feet contains much chert like Member 5. Basal 5 feet full of internal casts
of a Pentamerid shell. 145 feet.
. Massive, cream-colored or brownish dolomite, medium fine-grained,
weathering to smooth surfaces. 15 feet.
. Like Member 8. 138 feet.
. Like Member 4. Basal 20 feet crowded with casts of Pentamerids and
corals. 112 feet.
Marked Disconformity
MIDDLE AND UPPER ORDOVICIAN
Richmond (Upper Bighorn, Upper Fish Haven)
. White to light-gray or buffish, dense to coarsely crystalline dolomite, mostly
thin-bedded. Less resistant than Member 8. Cherty and very fossilifer-
ous at Goose Creek Ridge. M.T., 68 feet, Goose Creek Ridge.
On
PALEOZOIC STRATIGRAPHY OF ROCKY MOUNTAINS 127
. Cliff-making, massive dolomite, locally with algal structure. Light-buff
to brownish (dark-brown at Blacksmith Fork). In the Absaroka Range,
partly brecciated, mottled, brownish fragments in buff matrix. Sparingly
fossiliferous at Goose Creek Ridge. M.T., 270 feet, Blacksmith Fork.
. White to dark-drab or brown-gray, dense dolomite. Weak, breaking to
small angular fragments. Locally carries ostracods. M.T., 17 feet, Goose
Creek Ridge.
. Basal dolomitic breccia or conglomerate (not seen in Goose Creek Ridge
section), interleaving with, and giving place upward to, a massive, brown-
gray dolomite containing much calcite in seams and geodes, or to white or
drab, finely crystalline dolomite somewhat similar to Member 5. The
white dolomite carries corals at Goose Creek Ridge. M.T., 181 feet, Black-
smith Fork, where the two types of dolomite above described occur inter-
bedded.
Trenton (Lower Bighorn)
. Typically white, almost chalky, fine-grained dolomite, breaking into small
angular fragments with relatively smooth weathered surfaces. At Living-
ston Peak interbedded with more coarsely crystalline, brownish dolomite.
Locally carries ostracods. 56 feet, Dead Indian Creek (typical throughout).
M.T., 89 feet, Livingston Peak.
. Cliff-making, massive (rarely slabby), white to light-gray or light-buff dolo-
mite, medium to very coarsely crystalline, in many places in part with
brecciated structure. Sparingly fossiliferous in many localities. 147 feet, _
Labarge Mountain. M.T., 153 feet, Blacksmith Fork, where it is underlain ~
by 149 feet of massive, dark-brown dolomite, medium finely crystalline,
with many seams of calcite (both members included under Member 4 in the
diagrams). |
. Like Member 4, but thin-bedded or closely jointed, weak. M.T., 60 feet,
Labarge Mountain.
. (Recognized in the Crandall Creek and Dead Indian Creek sections only.)
Cream to buff, finely crystalline dolomite in 2-foot beds. Fossiliferous.
M.T., to feet, Dead Indian Creek.
. In Wyoming, developed as a white to buff or rose sandstone, mostly soft
and friable; fossiliferous, locally with fish remains. 29 feet (plus ?), Goose
Creek Ridge. Correlated with the Harding sandstone of Colorado. At
Manitou, very arkosic, and in part deeply stained (red and green); 47 feet
thick. Possibly contemporaneous with part of the Swan Peak quartzite.
Possibly of Black River rather than Trenton age.
Marked Disconformity
Swan Peak Quartzite
Five hundred feet thick, in the Randolph quadrangle, northern Utah.
Geneva sandstone or quartzite, north end of the Wasatch Mountains, Utah.
Eureka quartzite, 200 to 500 feet thick, east-central Nevada.
128 C. W. TOMLINSON
UPPER CAMBRIAN AND EARLY ORDOVICIAN
4. (Known at Blacksmith Fork only.) 640 feet. Includes the following types,
in descending order:
Thin-bedded, light-gray, finely crystalline dolomite; in the upper part
interbedded with olive shales and very fossiliferous. 206 feet. (Beekman-
town.)
Fossiliferous, dark blue-gray or blue-black, very finely crystalline dolo-
mite with much black chert throughout, and a 30-foot bed of black chert at
the base. 156 feet. (Beekmantown.)
Brownish-gray to bluish-gray, finely to coarsely caval bine dolomite in
1-foot to 5-foot beds. 278 feet.
3. Mostly thin-bedded to shaly, bluish-gray to brownish-gray, finely crystalline
dolomite and dolomitic shale, with many layers of flat-pebble limestone con-
glomerate. In part fossiliferous. M.T., 800 feet, Blacksmith Fork (lower
190 feet, Upper Cambrian, placed in the St. Charles formation by Walcott;
remainder Ordovician). ‘This member is represented throughout northern
and western Wyoming and southwestern Montana.
2. (Not differentiated except at Blacksmith Fork.) Finely to coarsely ceyeal
line, gray dolomite, mostly thin-bedded, weak; in part oolitic. 340 feet,
Blacksmith Fork.
1. Very massive, clifi-making, coarsely crystalline white dolomite, interbedded
near the top and bottom with minor beds of thinner-bedded gray dolomite.
M.T., 412 feet, Livingston Peak.
o. Mostly thin-bedded, white to light- and dark-gray, finely to coarsely crystal-
line dolomite, locally with much sandstone in lower part, and with inter-
bedded flat-pebble limestone conglomerate. In part fossiliferous. Lower
(barren) half called Middle Cambrian by Walcott at Blacksmith Fork
(upper part of the Nounan formation). M.T., 654 feet, Blacksmith Fork.
PALEONTOLOGICAL COLLECTIONS: ASSEMBLED LISTS
DEVONIAN
Members 5-11:
Actinostroma sp. Antler peak, Livingston Peak.
Alveolites goldfussi Billings. Livingston Peak.
Anplexus cf. hamiltoniae Hall. Antler Peak.
Blothrophyllum( 2) cf. cinctutum Davis. Antler Peak.
Zaphrentis(?) sp. Livingston Peak.
Atrypa missouriensis Miller. Livingston Peak.
Bryozoa. Antler Peak, Livingston Peak (2 sp.).
Member 3:
Airypa missouriensis Miller (fragmentary). Teton River.
Member 2:
Unidentifiable brachiopod and gastropod fragments. ‘Teton River.
Leperditia sp. Livingston Peak, Teton River (2 sp.).
PALEOZOIC STRATIGRAPHY OF ROCKY MOUNTAINS 129
SILURIAN (BLACKSMITH FORK ONLY)
Member 8, 10 feet below top:
Syringopora cf. verticillata Goldfuss.
Member 4, base:
Favosites sp.
Zaphrentis( ?) sp.
Conchidium knighti Sowerby.
Pentamerus ci. oblongus Sowerby.
Member 2:
Syringopora sp.
Member 1:
Favosites sp.
Conchidium knighti Sowerby.
Pentamerus cf. oblongus Sowerby.
UPPER ORDOVICIAN (RICHMOND)
Member 9. (Goose Creek Ridge only.)
Calapoecia( ?) cf. anticostiensis Billings.
C. cribriformis (Nicholson).
Favosites (n.sp. ?).
Halysites gracilis (Hall).
Streptelasma sp.
Crinoid fragments.
Rhynchotrema:sp.
Zygospira modesta(?) Hall
Orthoceras( ?) sp.
Bryozoa.
Member 8:
Calapoecia sp. Blacksmith Fork.
Streptelasma sp. Goose Creek Ridge, Blacksmith Fork.
Dalmanella ct. testudinaria (Hall) and hamburgensis (Walcott). Goose
Creek Ridge.
Orthoceras(?) sp. Goose Creek Ridge.
Member 7:
Crinoid fragments. Dead Indian Creek.
Dalmenella(?) sp. Crandall Creek, Teton River( ?).
Leperditia sp. Goose Creek Ridge, Crandall Creek, Teton River.
Member 6:
Calapoecia( ?) cf. anticostiensis Billings. Goose Creek Ridge.
Columnaria sp. Dead Indian Creek.
Halysites gracilis (Hall). Goose Creek Ridge.
Cf. Protarea richmondensis Foerste. Goose Creek Ridge.
Streptelasma sp. Goose Creek Ridge.
Pachydictya fenestelliformis(?) Nicholson. Goose Creek Ridge.
Rhinidictya cf. mutabilis (Ulrich). Goose Creek Ridge.
130 C. W. TOMLINSON
MIDDLE ORDOVICIAN
Member 4:
Spheroidal algae. Teton River.
Receptaculites oweni Hall. Livingston Peak.
Lichenaria cf. typa Winchell and Schuchert. Rattlesnake Mountain.
Columnaria alveolata Goldfuss. Rattlesnake Mountain, Dead Indian
Creek, Crandall Creek.
Halysites gracilis (Hall). Blacksmith Fork.
Streptelasma corniculum Hall. Goose Creek Ridge, Dead Indian Creek
(sp. ?), Crandall Creek. 6
A new cyathophyllid coral. Crandall Creek.
Member 2:
Receptaculites oweni Hall. Crandall Creek.
Halysites gracilis (Hall). Dead Indian Creek, Crandall Creek.
Streptelasma corniculum Hall. Crandall Creek.
Zygospira sp. Crandall Creek.
Clinoceras(?) sp. Crandall Creek.
Member 1, within 5 feet of top (Goose Creek only):
Lophospira sp.
Raphistoma( 2) sp.
Cyrtoceras(?) sp.
Orthoceras sp.
Receptaculites owent Hall.
EARLY ORDOVICIAN (BEEKMANTOWN) (BLACKSMITH FORK ONLY)
Member 4, near top:
A small Streptelasma-like coral.
A small cylindrical bryozoan.
Several species of orthid and strophomenoid brachiopods, including a
form probably identical with that called by Walcott! “‘Orthis testu-
dinaria,”’ from the Upper Pogonip; a form very similar to, and perhaps
identical with, the one called by White? “Strophomena fontinalis”’; and
a form which is probably identical with that called by Walcott “‘Orihis
perveta.”
Two species of low-spired gastropods.
Orthoceras sp. (small, annulated form).
1C. D. Walcott, ‘Paleontology of the Eureka District, Nevada,’ U.S. Geol.
Survey, Monographs, VIII (1884).
2C. A. White, “Invertebrate Paleontology: Report upon Geographical and
Geological Surveys West of the One Hundredth Meridian,” IV, Part 1 (1875), Engineer
Department, United States Army.
3 Op. cit.
PALEOZOIC STRATIGRAPHY OF ROCKY MOUNTAINS 131
Cf. Bathyurus? (Hystricurus?) tuberculatus Walcott. —
Ceraurus(?) sp. ;
An I[sotelus-like pygidium.
Member 4, 390 feet above base:
Lingula, 2 sp.
Two species of strophomenoid brachiopods; the same as the first two
of the three forms mentioned in the fauna collected near the top of
Member 4.
Orthoceras sp. (small, annulated form).
Asaphus(?) sp.
Ceraurus(?) sp.
UPPER CAMBRIAN
Member 3.
Billingsella coloradoensis(?) (Shumard). Antler Peak.
Eoorthis cf. remnicha Winchell. Teton River.
Obolella( ?) sp. Teton River.
Agnostus sp. Teton River.
Ptychoparia(?) sp. Antler Peak, Teton River( ?).
Trilobite fragments. Blacksmith Fork.
Member 2:
Unidentifiable organic (algal?) structures. Blacksmith Fork.
Member r:
Spheroidal algae, + inch to 2 inches in diameter. Blacksmith Fork.
. Member o, top (Blacksmith Fork only):
Billingsella coloradoensis (Shumard).
Lingulella manticula (White).
A gnostus.
Ptychoparia.
DISCONFORMITIES
In the following statement is listed the evidence pointing toward
discontinuity of sedimentation at the several horizons indicated.
8. Between Members 20 and 21 of the Devonian system
(between Devonian and Mississippian ?) :
a) Member 20B, a limestone, 361 feet thick at Blacksmith Fork, has no
- lithologically similar representative in any of the other sections, unless it be
a 10-foot limestone at Logan, Montana.
b) Member 21 rests on Member 19 at Crandall Creek and at Dead Indian
Creek.
c) At Labarge Mountain, Teton River, and Livingston Peak, according to
the writer’s interpretation, the Madison limestone rests directly on Member 109.
132 C. W. TOMLINSON
d) Where Members 20B and 21 are absent, the thickness of Member 19 is
variable, though not extraordinarily so.
e) Member 21 consists of clastic sediment, chiefly sandstone and black
or deeply stained shale.
f) Member 21 contains “‘a fauna which is different in most of its forms
from that of the lower members, and is more like that of the Madison lime-
stone’? which overlies it.
7. Between Members 18 and 19 of the Devonian system
(between the Jefferson limestone and the Three Forks formation):
a, Sharp lithologic change at this horizon.
b, Member 18 is moderately variable in thickness.
c, In the Teton River section the base of Member 109 contains nodules
of limonite. In the Crandall Creek section the same horizon is very deeply
iron-stained and carries small geodes of amorphous hematite.
It will be noted that this evidence is entirely circumstantial.
6. Between Members 14 and 15 of the Devonian system
(between the main and upper divisions of the Jefferson dolomite) :
a) There is a 16-foot sandstone at this horizon in the Labarge Mountain
section, and much sandstone in Member 15 in the Teton River section.
b) Members 11-14, with a maximum aggregate thickness of 550 feet, are
absent in the Teton River, Antler Peak, and Crandall Creek sections.
c) Members 12-14 are not known north of Labarge Mountain.
d) The Nevada limestone of eastern Nevada includes a lower and an upper
fossiliferous zone, separated by from 2,000 to 4,000 feet of barren beds. The
Jefferson fauna includes elements of both of the fossiliferous zones of the
Nevada, suggesting that the great thickness of the Nevada is due to the presence
of medial barren members which are not found in the Jefferson.?
e) In the Snowy Mountain section in Yellowstone Park, Weed describes
a 25-foot belt of limestone conglomerate at what may be this horizon.
5. Between Members 3 and 4 of the Devonian system (between
the basal and main divisions of the Jefferson dolomite):
a) There is a 10-foot bed of sandstone (Member 4) at this horizon in the
Labarge Mountain section, and a thinner bed of extremely sandy dolomite in
*W. P. Haynes, “‘The Fauna of the Upper Devonian of Montana, Part 2, The
Stratigraphy and the Brachiopoda,” Annals of the Carnegie Museum, X (1916), 27.
It is to be noted that Haynes reached the conclusion that, nevertheless, “there is no
sharp break in the record here” (zbid., p. 20).
2See discussion of ‘‘The Jefferson Dolomite.”
3W. H. Weed, “Geology of the Southern End of the Snowy Range,’ U.S. Geol.
Survey, Monographs, XXXII, Part 2 (1899), chap. vi, p. 206.
PALEOZOIC STRATIGRAPHY OF ROCKY MOUNTAINS — 133
the Teton River and Crandall Creek sections marks the base of the main divi-
sion of the Jefferson.
b) Member 3 was not recognized in the Antler Peak or Dead Indian Creek
sections, and in no known section north of Labarge Mountain does it have more
than a small fraction of its thickness in that locality.
c) No part of the basal division of the Jefferson was recognized in the Logan
section.
4. At the base of the Devonian system:
a) The Silurian system, 750 or more feet thick in northern Utah, is not
known in Wyoming or in Montana.
b) The Upper Bighorn dolomite, which underlies the Devonian in western
Wyoming and in part of southern Montana, varies greatly in thickness in that
region.
c) Members 8 and g of the Ordovician system, which elsewhere attain a
net thickness of 270 feet, are not found in the Teton River section.
d) In Montana, west of the Gallatin Range, the Devonian system rests
on Cambrian strata.
To summarize points (a) to (d), the Devonian system in different parts of
the central Rocky Mountain region rests upon the Silurian, the Ordovician,
and the Cambrian, respectively.
e) In the Teton River section an erosion surface marks the base of the
Darby formation.
f) Inthe Crandall Creek section there is at this horizon a thin conglomerate
of small rounded pebbles of dolomite in a matrix of laminated, iron-stained
shale, overlain by thin lenses of very deeply iron-stained shale. The basal
bed varies notably in thickness within a few yards along the strike, showing
that it was deposited upon an irregular surface.
It is obvious that there is a hiatus at the base of the Devonian
system wherever the Silurian is missing.
3. At the base of the Silurian system:
a) In the Blacksmith Fork section there is an indubitable erosional dis-
conformity at this horizon.
b) The brachiopod fauna above this disconformity is wholly different from
any found below it.
2. Between Members 5 and 6 of the Ordovician system (between
the Trenton series and the Richmond series, Lower and Upper
Bighorn, Lower and Upper Fish Haven, Lower Bighorn and Leigh):
a) In the Crandall Creek, Dead Indian Creek, and Teton River sections
there is a well-marked erosional unconformity at this horizon. (Also at the
t Eliot Blackwelder, personal note.
134 C. W. TOMLINSON
base of the Fish Haven dolomite in the Randolph quadrangle, Utah; but
this may perhaps be at the base of the Trenton.)
b) In the Crandall Creek, Dead Indian Creek, and Teton River sections
there is at the base of the Upper Bighorn (or Leigh) a breccia or conglomerate
of dolomite pebbles in dolomitic or shaly matrix, up to several feet thick; and
similar conglomerate occurs at the base of the Upper Fish Haven in the Black-
smith Fork section.
c) Thin lenticular bands of deeply stained shale appear in and above the
conglomerate in the Dead Indian Creek section, and the matrix of the con-
glomerate is deeply iron-stained both there and in the Teton River section.
d) Member 5 is not known south or southwest of Cody, Wyoming.
e) A hiatus between Trenton and Richmond was inferred by Darton? in
the Bighorn Range from paleontological evidence.
1. At the base of the Middle Ordovician series (Lower Bighorn,
Lower Fish Haven):
a) Throughout Wyoming, Montana, and South Dakota, wherever the
Bighorn formation (or a formation correlated with it) exists it rests upon strata
which are classed as Cambrian, and which certainly in no case are younger than
Beekmantown.
b) The basal member of the Bighorn at several localities in Wyoming is
a sandstone, of variable thickness, suggesting deposition on an uneven surface.?
c) In the Dead Indian Creek section the base of the Bighorn is a slightly
irregular surface. °
d) Inthe Randolph quadrangle the Fish Haven dolomite (all of Richmond
age| ?]) rests disconformably on the Swan Peak quartzite.
e) The Swan Peak quartzite is entirely missing from the Blacksmith Fork
section, although it is several hundred feet thick a few miles northeast and a
few miles southwest of that locality.
f) In eastern Nevada the contact between the Lone Mountain limestone
and the Eureka quartzite is clearly an erosion surface. .
«N. H. Darton, “Description of the Bald Mountain and Dayton Quadrangles,”
Geol. Atlas U.S. (Folio 141, 1906), p. 4.
Cf. N. H. Darton, ‘‘A Résumé of the Ordovician Geology of the Northwest,”
Bull. Geol. Soc. Amer., XVII (1905), 547. _
[To be continued|
SOME FACTORS AFFECTING THE DEVELOPMENT
OF MUD-CRACKS'
E. M. KINDLE
Geological Survey of Canada, Ottawa
INTRODUCTION
Casual examination of the apparently erratic lines known vari-
ously as mud-cracks, sun-cracks, and shrinkage-cracks affords little
promise of results of interest from their systematic study. The
great geologic interest which these products of desiccation possess
in connection with the history of formations in which they occur
should nevertheless encourage the geologist to ascertain what effect
variation in the conditions under which they are formed will have
on the resulting kind or type of mud-crack. With the object of
ascertaining the nature and extent of the modification of the type
of mud-crack which may result from varying the conditions of its
formation, I have carried out the laboratory experiments described
below. ‘These have been planned with a view to discovering (a) the
relative effects of rapid and slow desiccation on the same mixture,
(6) what influence, if any, composition of the mud has upon the
mud-cracks, (c) the possibility of producing parallel mud-cracks,
and (d) the differences which distinguish saline from fresh-water
mud-cracks.
EXPERIMENTS
Two kinds of clay have been used. In experiments 1 and 3 a
mud was used which came from the bottom of Lake Ontario, at
a depth of 630 feet, and represented very fine-textured material.
The other experiments were made with blue marine clay of Pleis-
tocene age from the Ottawa valley near Ottawa. ‘This is also a
very fine-textured and tenacious clay.
Experiment 1.—Lake clay which was thoroughly mixed with
about 5 times its volume of fresh water was poured into two
«Published with the permission of the Director of the Geological Survey of
Canada.
135
136 E. M. KINDLE
porcelain vessels of the same shape and size, each being completely
filled. One of these was set in direct sunlight when the daily noon
temperature exceeded too’, and the other was kept in the shade
until the water had evaporated and mud-cracks had developed.
The specimen exposed to sunlight. developed mud-cracks on the
third day. The other vessel showed the first mud-cracks on the
eighth day. When completely dried, the mud in the two vessels
showed a very marked difference in the size and number of polygons
outlined by the mud-cracks developed. The sun-dried mud had
cracked into 6 irregular-sided polygons, while the same volume
of mud, which had been slowly dried in the shade during a period
about three times that given the direct sunlight specimen, showed
26 polygons. Rapid drying thus seems to produce comparatively
widely spaced mud-cracks, while slow desiccation gives closely
spaced mud-cracks. Some interesting incidental observations
were made in connection with this experiment on the tendency dis-
played by the very fine sand grains to segregate themselves from
the mass of the mud and to gather along the edges of the joints and
the margin of the vessels holding the mixtures. This segregation
of the sand.grains resulted in a ring of sand around the outer margin
of the mud where it came in contact with the sides of the vessel.
Along the sides of many of the mud-cracks the upper angle of the
polygon was edged by a continuous border of sand. On the lower
side of the polygon edges the sand showed no tendency to segregate.
This segregation of the sand along the edges of the mud-cracks
appeared to be dependent in part upon the extreme fineness of the
grains. An attempt to repeat this phase of the experiment by
adding sand of average fineness to mud which was desiccated in the
sun failed to show any segregation phenomena, presumably because
of the larger size of the sand grains used.
Experiment 2.—This experiment was designed to show what
effect variation in the composition of the mud used would have on
the character of the mud-cracks. Three parallel experiments were
carried out for this purpose. The fine-textured blue clay of
Pleistocene age from the Ottawa Valley was used. A mixture of
this clay with 3 quarts of water was divided into three equal parts.
To one of these (3a) was added 2 ounces of fine sand. The same
FACTORS AFFECTING DEVELOPMENT OF MUD-CRACKS 137
amount of powdered marl was added to the second (20), while the
third lot was left a clay and water mixture (2c). These three mix-
tures, representing sandy mud, marly mud, and clay mud, were
placed in three shallow pans for desiccation in the sun. The result-
ing mud-cracks show that the clay mud (2c) cracked into much
larger polygons (Fig. 1) than either the sandy or marly mud (Fig. 2).
The sandy mud (3a) developed more than three times as many
Fic. 1.—Normal fresh-water mud-cracks in blue-clay mud. #} natural size
polygons (Fig. 3) as the clay mud, while the marly mud showed
more than twice as many polygons as the clay mixture. The large
number and very angular course of many of the mud-cracks in
the sandy mixture are characteristic features which distinguish
this mixture from either of the other two.
Experiment 3.—A portion of the same clay mixture used in
experiment 1 was placed in a shallow pan 11 inches in diameter.
The water was allowed to evaporate slowly without exposure to
138
the sun.
E.M. KINDLE
Fic. 2.—Mud-cracks in mud composed of clay
and marl. 3 natural size.
One side of the pan was raised ;'; inch higher than the
opposite side, so that
near the end of evap-
oration the lower side
remained moist after
the upper side had be-
come quite dry, the
object being to see
what effect, if any, this
drying out of the mud
in a lateral direction
would have on the
character of the mud-
cracks.
this experiment is
shown in the photo-
graph (Fig. 4), which
was made before the
lower margin had en-
tirely dried, the moist portion being indicated by the darker part
near the base of the
picture. Instead of
the usual reticulated
mud-crack lines, most
of the mud split up
into a set of ribbon-
like strips averaging
4 inch in width and
having a length of 3 to
6 inches. The direc-
tion of the mud-cracks
which gave this ribbon-
like effect was trans-
verse to, and evidently
controlled by, the zone
separating the com-
pletely dried from the
Fic. 3.—Mud-cracks in sandy mud. j natural
size.
The result of
FACTORS AFFECTING DEVELOPMENT OF MUD-CRACKS 139
partially dried mud (see Fig. 4). The cracks developed with the
retreat of this zone away from the area which first dried. In a
small patch of this first dried section no mud-cracks formed.
This experiment shows that approximately parallel mud-cracks
may be developed by differential desiccation, and affords a clue
to the cause of certain kinds of joints which appear to be definable
as parallel mud-cracks of considerable vertical extent.
Fic. 4.—Mud-cracks cutting the mud into ribbon-like strips. 4 natural size
Experiment 4.—A 2-quart mixture of blue clay and water was
divided into two equal parts. A tablespoonful of salt was added
to one of these, and the other was left fresh. The two mixtures were
placed in shallow pans 93 inches in diameter and put in the sun for
evaporation and desiccation. Complete drying or desiccation of
the salt-water pan was finished on the eighth day after starting this
experiment. The first noted difference between the two pans was
the earlier drying out of the saline mixture. All the liquid water
had left the salt-water mixture at least a day before the fresh-water
mud had ceased to be a semi-liquid mass. The desiccation was
finished in a temperature of 110°.
T40 E. M. KINDLE
In the fresh-water mixture preliminary mud-cracks developed
on a dried-surface layer of the thickness of paper two days before
the mixture underneath had lost its semi-liquid character. The
earlier drying out of this surface layer of paper-like thinness retarded
the drying of the lower layers and led to the excessive curling of the
polygons as they were cut out by the developing mud-cracks. A
lot of closely curled pieces of sediment resembling shavings repre-
Fic. 5.—Desiccated fresh-water mud. {natural size
sented the final stage of the mud-crack development in the fresh-
water mixture (see Fig. 5). An interesting feature of this experi-
ment is the difference in color exhibited by the thin uppermost
layer, which had been directly exposed to the air and sun, and the
sediment below. The topmost film had a lead-gray color, while the
sediment below it showed a creamy-white color in no way resembling
the original blue clay. The general character of the mud-cracks
shown by this pan corresponds closely to those most commonly
met with in nature except in the extreme curling of the polygons.
The behavior of the saline mixture was markedly different from
that of the fresh-water one. Except for a crack extending round
FACTORS AFFECTING DEVELOPMENT OF MUD-CRACKS 141
the margin of the pan and separating the mud adhering to the side
from that on the bottom no regular mud-cracks developed until
a very late stage of: the desiccation. Instead of the usual familiar,
somewhat erratic, mud-crack lines seen on drying mud, a five-rayed
star-shaped figure (see upper right-hand quarter of Fig. 6) cutting to
the bottom of the sediment first appeared. A day later two other
figures developed, each having three lines of equal length and form-
Fic. 6.—Desiccated saline mud. The same quantity and kind of mud was used
as in Fig. 5 except that salt was added. Note that margins of polygons are curved
downward instead of upward as in the fresh-water mud-cracks shown in Figs. 1, 2, 3,
and 5. 4 natural size. :
ing at their junction angles of 120°. Simultaneously with the
development of the three-line figures the entire surface became
marked by small hexagonal polygons with a diameter ranging from
s inch to ;/; inch and giving it a honeycomb appearance. These
may be seen indistinctly on the left half of Fig. 6. These were not
sharply defined nor marked off by mud-crack fissures, but were
discernible through a slightly lighter color of the sediment along
the bounding lines, and in some cases by a slight deliquescence of
142 E. M. KINDLE
salt along these lines. These small polygons appear to represent
the convection cells of Benard,’ Dauzére,? and Sosman,3 and have
no direct relationship to mud-cracks. A few hours after the ap-
pearance of the triradiate figures some regular mud-cracks formed
in the median portion of the pan, cutting a limited area into rather
small polygons. Two of these mud-cracks were extensions of arms
of the three-line figures previously mentioned. Two days after the
desiccation appeared to have been completed, the remainder of the
surface cracked, after being removed from the sun, splitting the
entire surface into polygons. A noteworthy feature of these poly-
gons is downwarping of their margins and absence of lateral
shrinkage, which is in sharp contrast with the upwarping of the
sides and considerable shrinkage of polygons which formed from
the fresh-water mud. In fact, the saline mud showed as a whole
slight lateral expansion which was taken up by the arching upward
of the median portions of the polygons.
Considered from the standpoint of preservation as permanent
features in the strata of consolidated rocks, mud-cracks in saline
clays would have a rather poor chance of permanent preserva-
tion owing to their slight breadth. If preserved, they would be
quite inconspicuous as compared with ordinary mud-cracks. The
star-shaped figures, however, by reason of their broad and deeply
incised arms lend themselves well to preservation under natural
conditions of sedimentation and should be regarded, when found
on rock surfaces, as evidence of subaérial desiccation. This experi-
ment represents the behavior of highly saline mud such as would
be found on the shores of salt lakes or detached arms of the sea
rather than that of the muds ordinarily met with about the estuaries
of rivers, which have a much lower degree of salinity.
The salinity of ordinary estuarine mud was approximated in
another experiment. Sea water was used in still another. In all
these supplementary experiments, including a sample of mud having
less than the salinity of ordinary tide-flat mud, desiccation produced
tH. Benard, Les Tourbillons cellulaires dans une nap pe liquide, etc., thesis, Paris,
1901; Rev. gen. Sci., XI (1900), 1261-71, 1309-38.
2 C. Dauzére, Jour. physique, VI (1907), 892-99; VII (1908), 930-34; Assn. franc.
av. SCi., 1908, pp. 289-96.
3 Robert B. Sosman, ‘“‘Types of Prismatic Structure in Igneous Rocks,” Jour.
Geol., XXIV (1916), 219-24.
FACTORS AFFECTING DEVELOPMENT OF MUD-CRACKS 143
polygons in which the upper and lower surfaces were perfectly flat,
the edges showing no inclination either to warp up or down.
SUMMARY AND DISCUSSION
The experiments described above justify the following deduc-
tions: Rapid desiccation produces mud-cracks which are more
widely spaced than those produced by slow desiccation. In mud-
cracks occurring in rocks of the same or similar composition the
relative size of the resulting polygons would therefore serve as a
basis for inferring the relative temperatures under which they were
formed.
The composition and the resulting tenacity of the mud very
materially affects the spacing of the mud-cracks. The presence
of marly material or the addition of sand gives polygons which are
much smaller than those formed in clay mud (Figs. 1-3). In the
case of sandy mud a sufficient excess of sand entirely prevents the
formation of mud-cracks. Hence a bed of sand might be exposed
to subaérial conditions without furnishing mud-crack evidence
of the fact. Temperature and tenacity of the material are two
primary factors in controlling the spacing of mud-cracks.
Approximate parallelism of mud-cracks may result from zonal
drying of the mud. The parallelism seen in many systems of joint
structure may thus be duplicated under special conditions in
shrinkage-cracks in mud.
A high degree of salinity delays the formation of mud-cracks
and results in polygons in which the margins are inclined downward
(Fig. 6). These are in marked contrast to the polygons formed in
fresh-water mud, which dish upward, saucer-like (Figs. 1, 2, 3,
and 5). The polygons formed in mud with the salinity of ordinary
sea water warp neither upward nor downward at the margins,
but retain a flat surface. It should be pointed out here that the
marked differences observed in the experiments between the be-
havior of fresh-water, highly saline, and moderately saline muds
are not ordinarily so well marked in nature as the accompanying
illustrations might lead the reader to expect. The strong tendency,
as shown by the pan experiments, of fresh-water mud to warp
upward and of very saline mud to warp downward at the margins
of the polygons on cracking is modified and often neutralized by
T44 E. M. KINDLE
the tenacity of the mud, which on a mud flat may prevent the top-
most cracked layers from partially splitting away from the sub-
jacent layers, as they must do if this warping occurs. Clearly the
cohesion between the layers of mud is greater than that between
the smooth bottom of the pan and the mud in it. Observation of
sun-cracked fresh-water mud on the bottom of evanescent ponds
will show that the polygons warp upward or remain flat, according
to the tenacity of the mud. Where the tenacity of the mud is
slight, the saucer-shaped polygons are dominant.
In the case of fossil mud-cracks the geologist can make definite
deductions regarding the salinity of the original mud only where
there has been distinct upwarping or downwarping of the polygons.
Where the surface is flat, as is usually the case, lack of warping is
as likely to be due to the tenacity of the mud overcoming the warp-
ing influence of fresh water as to the normal influence of the salinity
of sea water. Where the polygons show a definite saucer-like
upwarp at the margins, however, the inference that they were
formed from fresh-water mud would be inevitable. I have de-
scribed! from bed A of the Mount Wissick section in New Bruns-
wick an example of this kind which in the light of these experiments
must be referred to continental or fresh-water conditions, although
I originally supposed it to have been formed on a tidal flat.
Fossil examples of the inverted-saucer type of polygon due to
the drying of very saline muds are apparently not very common.
Some peculiar structures in Silurian dolomite described by Gilbert?
and illustrated? by Kindle probably represent a phase of this
phenomenon. These curved plates in the Lockport dolomite near
Niagara Falls, which are probably the result of the desiccation
of highly saline sediments, were supposed by Hall to be of concre-
tionary origin. They immediately precede in the section a rock
series in which beds of gypsum and rock salt afford conclusive
evidence of the highly saline character of the sediments deposited a
little later.
t Geol. Surv., Can. Mus. Bull. 2, 1914, p. 37-
2“‘Undulations of Certain Layers of the Lockport Limestone” (Abstract),
Science, N.S., XXI (1905), 224.
3 U.S. Geol. Surv. Folio No. 190, 1914, p. 59, Pl. 24.
DOWNWARPING ALONG JOINT PLANES AT THE CLOSE
OF THE NIAGARAN AND ACADIAN?
LANCASTER D. BURLING
Canadian Geological Survey, Ottawa
The upper part of the Lockport dolomite exhibits a number of
interesting structural features. These are not confined to the
Lockport, but many of them have an important bearing on the
physical conditions during and immediately succeeding the forma-
tion of the Lockport and as such deserve attention. Among the
most striking of these are: (1) the huge vase-like residual masses
of the Lockport on Flowerpot Island? in the Bruce Peninsula of
Ontario; (2) the widespread doming of the strata which Kindle
has described’ as probably analogous in origin to the mud lumps
at the mouth of the Mississippi; (3) the arching of the strata form-
ing the uppermost or Eramosa beds of the Lockport dolomite in
Ontario, where it is ascribed in at least one instance* to the
presence of an underlying coral reef; (4) the burial of Devonian
rock and fossils in joints 18 feet below the present glaciated surface
of the Niagara limestone in Illinois;5 (5) the anticlinal arches which
characterize the Lockport in the Niagara region, and which Kindle
and Taylor® ascribe to local stresses of comparatively recent date;
(6) the ripple-mark and other sedimentation phenomena which
have been described so frequently;7 (7) the immediate super-
position above the Lockport of the Salina with its salt and gypsum;$
« Published by permission of the Deputy Minister of Mines.
2 Stauffer, Geol. Surv. of Canada, Guide Book No. 5, 1913, Pp. 75.
3 Amer. Jour. Sci., 4th Ser., XV (1903), 459-68.
4 Williams, Geol. Surv. of Canada, Museum Bull. No. 20, 1915, pp. 1-2. It should
be noted, however, that the horizon of the Eramosa beds is below the top of the Lock-
port as used, for example, by Kindle and Taylor in the Niagara Folio.
5 Weller, Jour. Geol., VII (1899), 483-88.
6 Geol. Ailas of the U.S., U.S. Geol. Survey, Niagara Folio (No. 190), 1913, p. 109.
7 Cf. Kindle, Geol. Mag., Dec. 6, I (1914), 158-61.
8 Cf. Grabau, Bull. Min. and Metal Soc. Amer., VI, No. 2 (1913), 33-44.
145
146 LANCASTER D. BURLING
and (8) the local downwarping along joint planes, to which this
paper is directed.
The downwarping of certain layers near the top of the Lockport
dolomite has been the subject of frequent reference, and illus-
trations of it have been copied and recopied. ‘The structure was
first described by Hall, who characterized it as concretionary.’
Chamberlin and Salisbury? first called attention to the fact that
the sag is along joint planes, an explanation which is concurred
in by Hobbs,3 but neither of these authors makes any comment
regarding the period of deformation. This was first treated by
Gilbert in 1905 in a paper of which we have only an abstract. After
describing the structures he says‘ that he is not satisfied with Hall’s
characterization of them as concretionary, but that they were
probably contemporaneous with the deposition of the strata and
not subsequent to it.
The phenomenon has been described for the following local-
ities: (1) Niagara limestone at Porter’s quarry, Niagara Falls;
(2) Niagara limestone, Cook’s quarry, near Lasalle, Niagara
County, New York;° (3) Lockport dolomite, Niagara Falls: (@) in
new railroad cutting; (b) in quarry 3 miles east of the city; and
(c) in water channels temporarily exposed at the Dufferin Islands
on the Canadian side;? (4) Lockport dolomite, Niagara Falls, old
quarry 15 miles east.®
The purpose of this paper is to show that such structures are
essentially contemporaneous with the deposition of the strata, that
they should be expected to occur where they do, that they are
important, and that they have formed under essentially similar
t Geol. New York, Part 4 (1843), p. 94, Fig. 30.
2 Geology (New York: Henry Holt & Co., 1904), I, 150, 151.
3 Earth Features and Their Meaning (New York: Macmillan, 1912), p. 224,
legend to Fig. 239.
4 Science, N.S., X XI (1905), 224.
5 Geol. New York, Part 4 (1843), p. 94, Fig. 30.
6 Chamberlin and Salisbury, Geology (New York: Henry Holt & Co., 1904), I,
Fig. 137, p. I5I.
7 Gilbert, Sczence, N.S., X XI (1905), 224.
8 Kindle and Taylor, Geol. Atlas of the U.S., U.S. Geol. Survey, Niagara Folio
(No. 190), 1913, Illus., III, Pl. XXIV.
DOWNWARPING AT CLOSE OF NIAGARAN AND ACADIAN 147
conditions in rocks of Cambrian age where the evidence as to their
early origin is conclusive.
In all of the Lockport localities that have been described the
warped surface directly underlies till or marine clays. During the
field season of 1915 a similar warped surface was discovered to be
characteristic of the uppermost beds of the Middle Cambrian in
British Columbia. These are dolomites and form the top of the
Eldon formation. The basal beds of the Upper Cambrian rest
directly upon this warped surface and, as if to yield further con-
firmation of the fact that the beds forming the top of the Eldon
suffered prolonged exposure to the air and that such a condition
continued during the period of deposition of the basal beds of the
overlying Bosworth formation, the latter is full of mud-cracks,
ripple-marks, and casts of salt crystals 2 inches or more in diameter.
The phenomenon was studied in the amphitheater north of
Castle Mountain, and there is here no question that the warping
was essentially contemporaneous with the deposition of the strata.
It is interesting to note that the only known occurrences of this
peculiar type of warped structure in both cases occur at the top of
a dolomite overlain by shales with salt crystals and evidences of
salinity. In the Lockport, Grabau' is of the opinion that the over-
lying Vernon suggests the accumulation of fine loess-like material,
chiefly as wind-blown dust. This is of interest in connection with
the theory of the eolian origin of the salt deposits of India discussed
by Holland and Christie.2 The bed immediately overlying the
warped structure at the top of the Eldon is almost a pure dolomite,
and the warped layers whose depressions it fills contain almost as
little calcium carbonate. Grabau’ describes the section between
the Lockport and the Salina as composed of a “stratum of thin-
layered bituminous accretionary limestone, forming flat, imbri-
cating, shell-like domes’ overlain by 2 feet of yellow impure
limestone, which is in turn succeeded by the green shale forming
the base of the Salina. In this paper on the early Paleozoic Delta
deposits of North America, Grabau goes into the physical conditions
t Bull. Geol. Soc. Amer., XXIV (1913), 490.
2 Rec. Geol. Survey India, XXXVIII, Part 2 (1909), 154-86.
3 Bull. Geol. Soc. Amer., XXIV (1913), 401.
148 LANCASTER D. BURLING
of Niagara time in great detail, but does not mention the warped
surfaces. Likewise Clarke and Ruedemann, in their memoir on
the Guelph,t do not mention these structures in their discussion
of the conditions of life and sedimentation during the prevalence of
this fauna. Writers agree, however, that there was a shallowing
of the sea near the close of Lockport sedimentation and a gradual
increase in its salinity and the magnesian content of its waters.
Arguing from the extraordinary thickness of the Guelph mollus-
can shells, Kindle and Taylor? postulate the subjection of the
bottom of the Silurian sea at this time to intense wave-action.
Calvin’ accepts the theory which had already been suggested by
Hall that ‘at the close of the Niagara huge mounds and ridges were
built on the bottom of the shallow Silurian sea, in part by the
accumulation zz situ of corals, crinoids, and molluscous shells, and
in part by the drift of calcareous sediments under strong currents.”
Tam inclined to the opinion, and this appears to be corroborated
by the position and physical character of the sediments involved,
that the sagging of these beds, both those in the Eldon of British
Columbia and those in the Lockport of New York, was largely
caused by the gentle scour of water at a time soon after deposition.
In each case the horizon of the warped structures is the locus of
pronounced changes in the paleontologic record. In the Lockport
the time was one of a prolonged emergence and marked the close
of the Niagaran; in the Eldon it marks the close of the Middle
Cambrian or Acadian and doubtless indicates a similar period of
emergence at that time. Walcott, who did not, however, have the
advantage of having seen the salt crystals of the Bosworth forma-
tion, says: ‘‘It is difficult to resist the conclusion that the 268 feet
of shales forming the base of the Bosworth Upper Cambrian section
were deposited in fresh or brackish water or on a river flood plain
or delta such as Barrell describes so graphically in his studies of
the Geological Importance of Sedimentation.”* ‘The warped structure
at the top of the Middle Cambrian in British Columbia has been
t New York State Museum, Mem. No. 5 (1903), 114-21.
2 Geol. Atlas of the U.S., U.S. Geol. Survey, Niagara Folio (No. 190), 1914, p. 116.
3 Geol. Survey Iowa, Rept., 1896, p. 120.
4 Problems of American Geology (Yale, 1915), p. 185.
DOWNWARPING AT CLOSE OF NIAGARAN AND ACADIAN 149
observed, not only in the Castle Mountain section along the
Canadian Pacific Railway, but in the Mount Robson section of the
Grand Trunk Pacific 200 miles to the northwest. In each case
it is followed by reddish-purple, green, and yellow shales, with
ripple-marks, mud-cracks, and casts of salt crystals.
Salt crystals, previously known only from pre-Cambrian and
post-Ordovician rocks, have been assumed, and rightly, to indicate
arid conditions and more or less emergence. ‘The peculiar type of
downwarping described appears to be the natural result of the
quiet subaérial exposure of recently consolidated dolomitic lime-
stones under conditions of aridity. Whether or not the down-
warping described is necessarily contemporaneous with aridity, it
is certainly a feature due to subaérial exposure, and the probabili-
ties are in favor of its formation at the time of the deposition of
the strata rather than subsequently. This necessarily involves the
assumption that the joints along whose channels the solution was
localized came into existence very soon after the deposition of the
strata and were relatively contemporaneous with the deposition
of the beds. The evidence as to the early origin of the warped
structures is so conclusive that we are justified in disregarding
such coincidences as the immediate superposition, where so far
discovered, of tills and clays upon the warped Lockport and such
inferences as that, since the solution took place along joint planes,
it must be comparatively recent. Have we not here rather a slight
measure of the duration of the time-break in the deposition between
the Middle and Upper Cambrian, and between the Niagaran and
the Cayugan, and are we not justified in bearing in mind the
principle of relatively contemporaneous consolidation and joint-
ing in rocks ?
‘THE WESTERN INTERIOR GEOSYNCLINE AND ITS
BEARING ON THE ORIGIN AND DISTRIBUTION
OF THE COAL MEASURES’
FRANCIS M. VAN TUYL
University of Illinois
While occupied with a study of the stratigraphy of the Missis-
sippian formations of Iowa for the Iowa Geological Survey the
writer has been attracted by the regular and nearly uniform gentle
dip of these formations to the southwestward, and he has recently
attempted to ascertain the age and significance of the deformation
which gave rise to this. Investigation soon showed that the tilting
was related to deformation over a wide area in southern Iowa, south-
eastern Nebraska, eastern Kansas, and northwestern Missouri
which outlined a great southwestwardly pitching geosyncline in
which the Coal Measures of the western interior coal field were
deposited. The evidence is clearly in favor of the view that the
movement took place, in part at least, during pre-Pennsylvanian
time, as shown by the fact that the Coal Measures now rest upon
trunkated Mississippian formations successively younger in age
toward the southwest. This belted arrangement of the Mississip-
pian deposits beneath the Coal Measures cannot be accounted for
on the assumption that the distribution of the former is original,
since the beds often consist entirely of nearly pure limestone up to
their very boundaries and show no indications of shore facies.
That the geosyncline was shallow in early Pennsylvanian time
is indicated by the fact that the maximum known thickness of the
deposits of the Cherokee stage, which probably represents the time
of greatest sea extension in this basin during the Pennsylvanian, is
only 712 feet. At the present time, however, it attains a known
depth of approximately 2,400 feet at McFarland, Kansas, and
future drill records will probably show it to be considerably deeper
than this to the southwest. The deepening is believed to have
t Published with the permission of the Director of the Iowa Geological Survey.
150
THE WESTERN INTERIOR GEOSYNCLINE 151
been brought about in part by subsidence during the post-Cherokee
stages of the Pennsylvanian and in part by post-Pennsylvanian
deformation. The data are not sufficient at present to warrant an
estimate of the relative importance of each of the two. There is
evidence that the original outlines and relations of the basin have
been considerably modified by these later readjustments.
The magnitude and significance of the basin have been demon-
strated by the construction of 1oo-foot contours on the base of the
Coal Measures, from data furnished by the reports of the state geo-
logical surveys of Iowa, Missouri, and Kansas (Fig. 1). Contours
showing the altitude of the base of the Coal Measures in Missouri
have already been drawn by Hinds and Greene,’ and these have been
copied directly. Norton has also drawn a similar contour map for
the southwestern and south-central portions of Iowa,? and this has
been adopted with little modification.
The presence of this basin not only explains the great dis-
similarity between the Coal Measures of this field and those of the
eastern interior field, which were undoubtedly deposited in a dis-
tinct basin, but also explains the belted arrangement of the out-
crops of the Pennsylvanian formations in Iowa, Missouri, and
Kansas, where the younger members are approximately confined
to the center of the basin, progressively older ones being exposed
toward its margins.
There can be no doubt that this geosyncline exerted an impor-
tant influence on sedimentation in this region during the Pennsyl-
vanian. The work of Hinds and Greene in Missouri has furnished
valuable data bearing on this point. Referring to the Cherokee
deposits of that state they say:
The Cherokee sea, advancing from the west or southwest, first invaded
Missouri in the vicinity of Forest City, Holt County, and soon extended north-
east as a long shallow arm through Worth, Harrison, and Mercer counties into
Iowa. When about 150 feet of Cherokee sediments had been laid down the
arm had broadened out to the southeast so as to embrace Buchanan and Platte
counties, and a short time later Clay, Jackson, and Livingston counties. After
the deposition of nearly 400 feet of material in the Forest City area the sea
covered practically all of the western tier of counties, except Atchison, and
* Mo. Bur. Geol. and Mines, Vol. XIII, 2d Ser. (1915), Pl. 25.
2 Towa Geol. Survey, XXI (1912), 1101.
FRANCIS M. VAN TUYL
152
So SS 6 SS 0 Or
--—_——
SHIN 40 31VOS
.SBINSEAW |B0D,, j0 AJEpunog eyewjxoiddy
s)|sodap jauueyo
aNnaoa1
NOSTAN 'S’M CONV “JANL NVA ‘Kid Ad NMYUC
ANITONASOAD AOIMGLNI NYILSAM
ONIA\OHS
dVW
THE WESTERN INTERIOR GEOSYNCLINE 153
soon extended eastward into Henry, Johnson, Lafayette, Roy, Carroll, Linn,
Putnam, and Adair. When the Bevier coal bed was formed, near the beginning
of Allegheny time and after the deposition of 580 feet of material at Forest
City, some sedimentation had already taken place in all the region now occu-
pied by the main body of the Pennsylvanian and a fairly large area in which
there are now only small patches remaining. The land area had been reduced
to an island in southeastern Missouri, with a peninsula projecting into Pike
and neighboring counties and a small part of a northern land mass in the
extreme northwestern corner of the state. The western sea continued to
advance eastward while an eastern sea occupying most of Illinois advanced
westward. Probably by the end of Cherokee time the two seas had joined,
submerging practically all of northern Missouri and possibly nearly all of
southern Missouri also. No deposition appears to have taken place at this
time in the extreme northwestern corner of the state, for the Nebraska City
drilling shows less than 100 feet of Des Moines strata, probably of Pleas-
anton age."
That these authors are justified in this conclusion is shown by a
study of the thicknesses of the Cherokee in Missouri as listed by
them.? Thus at Forest City in Holt County the thickness is 712
feet, while its average thickness in the counties to the eastward
becomes successively less and less, viz., Livingston 450, Linn 260—
310, Macon 175, Audrain 75. To the southeastward a similar
relationship is shown, thus: Buchanan 530, Platte 555, Clay 460,
Jackson 430, and Johnson 220-350 (see Fig. 2).
The influence of the basin upon the thickness and character
of the post-Cherokee stages of the Pennsylvanian is not so obvious.
The writer, after a careful study of all the available data, including
deep-well and drill records from various parts of the area, has not
been able to find any consistent variation in the thickness and litho-
logic character of the formations in tracing them from the center of
the basin toward its margins. Nevertheless,the, lack of relation
between these deposits and those of the Illinois field suggests that
the basin persisted and that its gradual though interrupted sub-
sidence made possible the deposition of the Pennsylvanian forma-
tions of this province. It is believed that the original relations
have been masked in large part by disconformities within the Coal
Measures, several of which have been recognized, and by post-
Pennsylvanian erosion, which has almost entirely removed the mar-
ginal facies of all the formations younger than the Cherokee.
1 Op. cit., p. 2009. 2 Op. cit., pp. 39-40.
154 FRANCIS M. VAN TUYL
SIGNIFICANCE OF THE CHANNEL DEPOSITS
The channel deposits in the Coal Measures of Iowa and Mis-
souri are very interesting in this connection in that they furnish
a —[pacrisen files emery
Gentry» afl Adair i:
ales T00 -- fn
ey esas oes == col ga
pa
frcrinnel nts) | | Hitter aaa! lev .Sectland. Clark:
4
jean ite
An eo | al infor)
abe z 5 (i Rudrain :
ioward * 75 an
132 /soone “=.
Jackson, 3230 | Re ntcrel ee Se | PAGE
430 —_— df: oi laway.
s 8 + }108 $
Johnson. « obi Ree
1 Coors 220-380 ae
o 340 | G
Henvy | =
Bates esses e
p225-370 | Peet
i .
inex eC ees Sr ee
Fic. 2.—Map showing variations in thickness of Cherokee formation in Missouri.
The influence of the basin is shown when this map is compared with that of Fig. 1.
corroborative evidence of the persistence of the basin in post-
Cherokee time. With reference to such deposits in Missouri
Hinds and Greene say:
Among the most unique features of the Missouri Pennsylvanian are two
long, narrow channels filled with sandstone and shale which have been eroded
in Cherokee, Henrietta, and some Pleasanton strata in Johnson, Lafayette,
Randolph, and other counties. Remnants of other channels have also been
found in many parts of the Pennsylvanian area, and many more probably remain
to be discovered as the net of detailed geologic work is spread over the state.
The channels are of great scientific interest, for they must have been formed
during an interval of more or less widespread emergence and erosion during
THE WESTERN INTERIOR GEOSYNCLINE 155
Pennsylvanian time. If, as suspected, the channel deposits are contempo-
raneous with certain sandstones and conglomerates of late Pleasanton age in
north-central Missouri, this erosion interval occurred before the beginning of
the Missouri epoch.
The location and trend of the channels are shown on the accom-
panying map (see Fig. 1). The east-west channel has been desig-
nated the “‘ Warrensburg” and the north-south one the “‘ Moberly.”’
In describing these deposits Hinds and Greene state that—
the Warrensburg sandstone fills a channel about 50 miles long, extending from
north of Lewis Station, Henry County, northward to the north bluffs of Mis-
souri River. The sandstone belt, as at present exposed, has an average width
oi two miles, but just south of the Missouri widens to six miles... . .
The Moberly channel extends from South of Madison, in Monroe County,
west to Chariton River south of Salisbury. Its length is nearly 40 miles and
its average width less than 3 miles. ‘The maximum depth shown in drill records
is about 200 feet.?
With regard to the nature of the streams which gave rise to the
channels the same authors express themselves as follows:
It is believed that the Warrensburg channel was made by water flowing
from higher country on the Ozark dome, bringing with it sands and muds
derived largely from early Pennsylvanian sediments. The Warrensburg stream
was joined when it reached the present site of Missouri by the Moberly River
descending westward from an Ozark peninsula in northeastern Missouri, and
the united streams continued northward or northwestward to the open sea.3
Referring to the channel deposits presumably of a similar age
in Iowa, Hinds and Greene say:
The Red Rock sandstone of Marion and Jasper counties, Iowa, lies in a
channel 2} to 3 miles wide that has been traced for 27 miles from Eagle Rock
northeastward. This sandstone has a maximum thickness of 100 feet and has
all the characteristics of the Warrensburg and Moberly sandstones.4
It will be noted that the trend of these old channels, both in
Missouri and in Iowa, indicates that the drainage development
during the temporary uplift in late Des Moines time was influenced
by the geosyncline.
tOp. cit., p. OI. 3 [bid., p. 93.
2 [bid., pp. 95 and 97. 4 Ibid., p. 94.
156 FRANCIS M. VAN TUYL
POST-PENNSYLVANIAN HISTORY OF THE BASIN
Subsequent to the deposition of the Paleozoic the Mississippi
Valley region was uplifted. It seems probable that the geosyncline
was deepened somewhat at this time and that secondary folds were
developed. But the fact that the regularity of the basin has not
been appreciably interfered with indicates that this secondary
folding was not of great importance. Following this uplift the
region underwent peneplanation. That the deposits occupying the
geosyncline were peneplained by the beginning of the Upper Creta-
ceous Is shown by the fact that the basal deposits of this age rest
upon the beveled edges of the dipping formations, ranging in age
from the Lower Mississippian on the marign of the basin in Iowa
to the Permian in Kansas, at approximately the same elevation
everywhere. ;
There is no evidence of further movement of the geosyncline
during or since the Cretaceous apart from the regional uplift
which brought the area to its present level. The present course of
the Missouri River across the basin was obviously taken some time
after the development of the peneplain.
A DECIMAL GROUPING OF THE PLAGIOCLASES!
F, C. CALKINS
United States Geological Survey, Washington, D.C.
The accompanying diagram, which represents about half of the
methods according to which the soda-lime feldspars have been
‘grouped, reveals a surprising diversity of usage. Consistency,
which undeniably would be of some advantage in the long run,
is not likely to obtain until it is demonstrated that some particular
scheme is better than all others. That this has not been done is
perhaps because it has not been attempted; for, once the matter is
given any critical attention, the most convenient and _ logical
adaptation of current nomenclature seems rather easy to find.
The ideal plan should, first, accord with the modern doctrine
that the plagioclases form a continuous series; schemes that
imply a limited number of compounds (Nos. 1-3) must, therefore,
be rejected.
Secondly, the intervals that separate the species in certain plans
(Nos. 5, 8, 9) are needless, even when they imply no discontinuity.
They then result in eleven-fold instead of sixfold division, com-
pound names, like “andesine-labradorite,” being applied to the
intervals between the main species; but six terms, with qualifying
adjectives, will suffice for as close discrimination as is worth while,
and the awkward compound names may be reserved for feldspars
that lie virtually at the junction of two species.
Finally, the division should be regular. The question arises here
whether the names “albite”’ and “‘anorthite”’ shall be applied only
to the pure soda and pure lime feldspars respectively, as in Zirkel’s
quite regular plan (No. 7), or whether they shall denote a certain
range of composition. As absolutely pure end terms are mere
abstractions, the answer to this question amounts to a choice
between a fourfold and a sixfold division. It is only following
universal practice to prefer the sixfold one. Now, there are but
t Published by permission of the Director of the United States Geological Survey.
157
158
F.C. CAEKINS
two simple ways of dividing regularly a series of six parts: either
to give all parts an equal range, or to give the end parts half the
The latter plan
range of the others, which are mutually equal.
GROUPING OF SODA-LIME FELDSPARS BY VARIOUS AUTHORS
Aloite Oligoclase Andesine Labradorite Bytownite Anorthite
a ml (aa
An.% 9 10 ZUM ad 40 50 60 70 80 90 ~—«100
|. Fouque’ 1879 4
ae ; q % 9
Levy
2. Lév 1888
Lacroix
Starx 1897 & 99 OO 18 O0 6 60 6 99 04 4
4 Tschermak 1865 PZZZZZZZ ZZ AANAAANAARAN
S.Rosenbusch 1885 i (ZZZZZZZZZZZ) ‘MM ——— NING) TEED
6.Dana 1892@) ZAM ——$<$— SANS
7. Zirkel 1894 @&ZZAZZ7ZZZZZZ ZAM ANANAAAANIANNAN
8.Rosenbusch 1905 = cee 0 ————— A SSNS) —_
9, Johannsen 1908 e VIZ/STZZZZZ oy RSs =
la. Average of 4-9 ean) (CLZZLZZZEZD (SSSSSSSS9 |
I]. Proposed Plan ILE AN QAR
An% 0 1D 0 30 40 a GO 70 &0 90 100
. F. Fouqué and A. Michel Lévy, Minéralogie micrographique, 1879.
. A. Michel Lévy and A. Lacroix, Les Minéraux des roches, 1888.
Fic. 1.—Sources of plans illustrated
. A. Lacroix, Minéralogie de la France et de ses colonies, II (1897), 130.
. H. Rosenbusch, Mikroskopische Physiographie, etc., 2d ed., 1885 (‘‘nach Tschermak’’).
I
2
3
4. G. Tschermak, ‘‘Die Feldspathgruppe,’’ Sitzungsberichte d. K. Akad. Wien, L (186s), 566.
5
6
. J.D. and E. S. Dana, System of Mineralogy, 6th ed., 1892; also used by J. P. Iddings, Rock Minerals,
1906, and by F. W. Clarke, “Data of Geochemistry,”’ Bull. 330, U.S. Geol. Survey, 1908.
. F. Zirkel, Lehrbuch der Petrographie, 2d ed., 1894.
. A. Johannsen, Determination of Rock-forming Minerals, 1908.
. Average of 4-9.
tr. Decimal grouping.
sit
8. H. Rosenbusch, Mikroskopische Physiographie, 4th ed., 1905, Bd. I, 348.
9
°
DECIMAL GROUPING OF THE PLAGIOCLASES 159
seems the more logical as well as the more accordant with usage.
It leaves equal spaces between the several types; for, if the most
typical andesine is average andesine, the most typical albite is
pure albite.
It is therefore proposed that the divisions be placed, as in
Diagram 11, where the ratios of anorthite to albite are ;'y, 7%, 74,
and ;°-
The questions of priority and of average practice have thus far
been left in the background. If priority determined preference,
Tschermak’s plan (No. 4) should be preferred; and the belief
seems general that his plan is most in use. But, rather oddly,
the scheme that currently passes for Tschermak’s is a modification
thereof by Rosenbusch (No. 5). The scheme proposed in the
present note resembles Tschermak’s more closely than does any
other. Still more closely does it resemble the scheme (No. 10)
deduced by averaging the ranges of species in Diagrams 4 to 9.
The slightest alteration that will regularize this “‘average”’ plan
and close its gaps produces the decimal grouping.
A decimal grouping goes naturally with centesimal symbols,
of which the most-used form is Ab,Anjoon (e.g., AbsAneo, or
Abz;sAn;;); these, moreover, present certain practical advantages.
They give, more quickly than those like Ab.An, and Ab,An, a
definite idea of relative composition—which is merely saying that
decimals are more easily subtracted than common fractions. The
decimal co-ordinates, too, upon which extinction-angle curves are
plotted, indicate the composition corresponding to a given angle
in percentages, which it is a needless trouble to reduce to a frac-
tion of a small denominator. Since in such curves the anorthite
increases toward the right, it is by the percentage of anorthite,
rather than by that of albite, that the composition is naturally
measured. Therefore, a symbol such as An 60%, or An.o, which
indicates this percentage alone, conveys the essential informa-
tion more economically than the symbol Ab,Ang; the former’s
greater convenience, however, is possibly outweighed by the greater
currency of the latter.
STUDIES FOR STUDENTS
A CLASSIFICATION OF BRECCIAS
W. H. NORTON
Cornell College, Iowa
Few geologic structures so lend themselves to diverse inter-
pretations as the beds of broken rock called breccia. Example
after example might be cited of breccias of Europe and America
which have been differently explained by different students during
the last half-century and as to whose origin no consensus is yet
attained. This diversity of opinion seems partly due to the large
number of processes by which rocks are broken up, assembled,
and cemented into breccia, and to the fact that breccias may offer
no very obvious and indisputable evidences of the method of their
making. Diagnosis generally requires the use of multiple working
hypotheses and may proceed chiefly by the process of elimination.
For this reason a genetic classification which the writer has pre-
pared in connection with a field study of certain breccias affecting
the Wapsipinicon stage of the Devonian of Iowa may prove of
interest to students of these structures.
The diagnosis of a breccia requires the close observation of its
most intimate characteristics as well as of its associations with the
adjacent rocks. The matrix may be like or unlike the fragments
lithologically. It may be a chemical precipitate, a sedimentary
deposit, or the detritus of attrition. In volume it may be greater
or less than the fragments—interstitial, merely filling the spaces
between the fragments closely packed, or preponderant, forming
the larger part of the rock-mass in which the fragments are sporadic.
The fragments may be of any size, from huge blocks down
to chipstone. Lithologically they may be similar or dissimilar,
according as they result from the fragmentation of a homogeneous
rock-mass or from that of heterogeneous beds. They may be
sharply angular, more or less rounded by attrition in earth move-
ments, or even in part water-worn and approaching a conglomerate.
160
STUDIES FOR STUDENTS 161
They may be local in derivation, produced by the breaking up
in situ of a terrane, or they may have suffered transportation from
distant sources. They may be simple or of complex and brecciated
structure, the result of an earlier brecciation.
Breccia may form a mass entirely destitute of planes of bedding.
When bedded in a distinct stratum it may be classified as endo-
stratic.
A crackle breccia, representing incipient brecciation, is one
whose fragments are parted by planes of fission and have suffered
little or no relative displacement. The fragments match along
their apposed sides. The matrix is confined to the seams and is
commonly a chemical deposit.
A mosaic breccia is one whose fragments have been largely but
not wholly disjointed and displaced. The system of continuous
cracks of the crackle breccia has been destroyed, but more or less
of the fragments still match along adjacent surfaces and show that
they are consanguineous parts of once unbroken laminae or larger
beds. The term suggested is not a happy one, yet these breccias
may recall some ill-preserved mosaics of ancient ruins. The
matrix is confined to the seams and to the wider and irregularly
shaped interstices.
A rubble breccia is one in which no matching fragments are
parted by initial planes of rupture. The fragments are close-set
and in touch.
A breccia of sporadic fragments is one in which the fragments
are imbedded in a preponderant matrix like plums in a pudding.
It recalls the term ‘“‘plum-cake rock,” applied by the quarrymen
of North England to the breccias of their region. In his study of
the Permian Midland breccias of England, King" distinguishes
an endostratic breccia of this class by terming it “breccia sand-
stone,’ thus emphasizing the bedded matrix rock.
Breccia may be fossiliferous. Fossils may be restricted to the
fragments or to the matrix or may be found in both. Fossils of
the matrix may themselves be fragments. Brecciation may be
practically contemporaneous with the involved deposits. Such
are the intraformational breccias of Walcott. In certain classes
tW. W. King, “Permian Conglomerates of the Lower Severn Basin,” Quar.
Jour. Geol. Soc. London, LV (1899), 105.
162 STUDIES FOR STUDENTS
of breccia the matrix is of the same age as the fragments. In
other classes it is younger, and in still others it may even be supplied
by an older bed.
Breccia may be of slight extent and evidently due to local
causes. On the other hand, some of the most perplexing breccias
are regional and demand causes equally widespread in operation.
According to the conditions under which they accumulate,
breccias may be classified as subaérial, subaqueous, or endolithic,
formed within the lithosphere, the earth’s crust.
In classifying breccias genetically it will be remembered that
the making of a breccia involves, or may involve, three distinct
processes—fragmentation, assemblage of the fragments, and
cementation by the introduction of the matrix. In any given
breccia all of these processes may belong to the same or to different
categories. ‘The genetic classification may have in view either
fragmentation, as in crush breccias, or assemblage, as in several
other types.
Subaérial breccias, in which both fragmentation and assemblage
are above ground, may be classified as follows:
Residual breccia Bajada breccia
Talus breccia Glacial breccia
Rock-glacier breccia Volcanic breccia
Landslide breccia
Residual breccia.—This type is formed of the angular débris of
the waste mantle. It has been designated as ‘‘basal breccia,’
since it corresponds in position to a basal conglomerate. But all
subaérial breccias covered and preserved by the deposits of a
transgressing sea may correspond -equally well to basal conglom-
erates, so that some designation seems preferable which suggests
the residual origin of this specific kind.
Residual breccias develop especially on Karst topographies
where the limestone of the country rock contains cherty beds.
Under long denudation the surface with its characteristic sink-
holes and closed valleys comes to be covered in places to some depth
with breccia of sharp-edged chips of flint, and this by submergence
may be incorporated into the sequence of the geologic formations.
xW.S. Smith and C. E. Siebenthal, U.S. Geol. Surv., Geol. Atlas, Joplin Folio
(148), p. 9.
STUDIES FOR STUDENTS 163
The residual breccia of the Joplin district of Missouri has ‘been
described by Winslow,’ Smith and Siebenthal,? and others. The
breccia-producing rocks are Mississippian limestones rich in chert.
At the close of the Mississippian the region was uplifted and, under
long subaérial erosion, developed a typical Karst topography
mantled with angular cherty waste. During early Pennsylvanian
times the area was deeply submerged; the Carboniferous sea
transgressed, but without either leveling the relief or assembling
and rounding the residual cherts. The matrix is largely supplied
by the sea-clays of the Cherokee formation. ‘The residual breccia
thus formed suffered further changes. Solution of the underlying
limestone has caused the breccia to founder, producing complex
brecciation and intermingling blocks of the country rock. Ground-
water has also introduced a matrix of lead and zinc ores and jas-
peroid silica.
Talus breccia—Accumulating at the foot of cliffs, from frag-
ments broken off by frost and temperature changes, talus breccia
is a rubble of sharp-edged fragments, wedge-shaped in radial section,
and quite devoid of bedding. Stratification may be rudely simu-
lated, since slabs and other unequiaxed blocks creeping and sliding
down the slope come to rest with their longer axes parallel with
the surface. Owing to greater inertia, the larger blocks may
gather at the base. Small fragments prevail, but blocks of some
considerable size may be supplied by local sapping. The matrix
(the finer material derived from the weathering of the cliff and of
the talus) is interstitial and lithologically identical with the frag-
ments. Some of the matrix, however, may be foreign—dust and
sand brought in by wind, humus of a soil cover, and travertine
deposited by springs issuing from the cliff’s base.
Talus breccias are local and extremely limited in width. Even
under an arid climate favoring the perpetuation of cliffs, even in
the hamada desert, where long lines of marching cliffs form the
high risers for the broad steps of rock plateaus, talus accumulations
are rapidly consumed by deflation and are but a few rods wide.
The material is local and there is no zonal arrangement of fragments
of different lithologic kinds.
t Arthur Winslow, Missouri Geol. Surv., VII (1894), 464 f. HOW (Giicn 185),
164 STUDIES FOR STUDENTS
An example of a talus breccia is described by Wilson’ as occur-
ring at the base of a cliff of Keweenawan sandstones covered by
a sheet of ancient diabase.
Rock-glacier breccia.—A subspecies of talus breccia is that pro-
duced by rock glaciers, talus glaciers, or rock-streams, as they are
variously known.” Talus may be very rapidly produced where the
wedgework of frost is especially efficient, as in cold climates, on
the lofty walls of cirques composed of rock that favors fragmenta-
tion by close jointing, or other structures. Under the urge of its
own weight and that of the expansion of interstitial ice talus
creeps forward from the feeding cliff in long tongues of waste which
in shape somewhat resemble glaciers. Rock-glacier breccia differs
from talus breccia only in its greater extension normal to the cliff,
in the movement of the material to greater distances and with
very gentle slopes, and in the fact that unequiaxed blocks may be
expected to be found set at all angles, owing to the movement of
material en masse.
The famous limestone breccias of Gibraltar, described by Ramsay
and Geikie,? are attributed to this class. ‘These breccias occupy
wide tracts about the base of the Rock of Gibraltar and reach a
thickness of at least 1oo feet. Their slope in places does not
exceed two or three degrees. The fragments are almost invariably
quite angular. They vary in size from grit up to blocks 12 feet
or more in diameter, and are distributed without regard to size
and shape. They are no larger at the base of the cliffs than on the
outskirts of the formation. The matrix is earthy and cements
the breccia into firm rock. The authors cited attribute frag-
mentation to the work of frost under far severer climatic conditions
than now obtain. The assemblage of the breccia by gravity in
ordinary talus is negatived by the extent and slope and apparently
by the set of the fragments. The angularity of the fragments, their
size, and their lack of sorting preclude the theory of transportation
tA. W. G. Wilson, ‘‘Trap Sheets of the Lake Nipigon Basin,” Bull. Geol. Soc.
Am., XX, 207-9.
2 Whitman Cross and Ernest Howe, U. S. Geol. Surv., Geol. Atlas, Silverton
Folio, p. 25; S. R. Capps, “Rock Glaciers in Alaska,” Jour. Geol., XVIII, 359-75.
3A. C. Ramsay and James Geikie, Quar. Jour. Geol. Soc. London, XXXIV,
505-41, 1878. 7
STUDIES FOR STUDENTS 165
by torrential streams upon detrital slopes. It is concluded that
heavy trains of débris saturated by meltings of thick snows have
moved en masse down the steep slopes and out over the lower
grounds below. The junior author reaffirmed this theory in 1881
and compared the breccia directly with the talus glaciers of the
Rocky Mountains."
A matter perhaps less satisfactorily explained is the frequent
brecciated aspect of the undisplaced limestone of the Rock of
Gibraltar. In the words of the authors cited, ‘In many places
the Rock looks as if it had been smashed up zm situ, the broken
fragments having been subsequently consolidated by infiltration.”
The authors prove that the shattering is not due to faulting, and
attribute it to frost. The fractures recall, however, those of the
country rock of the landslide area of the Rico Mountains, referred
by Cross? to prehistoric earthquakes of exceptional violence.
Indeed, the features of the Gibraltar breccias as described by Ram-
say and Geikie are not inconsistent with an origin in landslides,
which also move out over gentle slopes and carry large fragments
to the outer limits.
Landslide breccta.—Breccias of this class owe fragmentation in
large measure, and assemblage wholly, to the force of gravity. In
rock-falls the movement is sudden and violent, and the rock is
shattered to pieces by successive impacts. Rock-fall breccia is a
chaotic mixture of blocks, large and small, set at all angles and
sharply angular. In the few seconds of the tremendous downrush
of perhaps millions of tons of rock, the fracture of the masses to
smaller and still smaller fragments and their crush to powder go on
so rapidly that comparatively little opportunity seems to be given
to the rounding of edges by attrition. The matrix consists of
chinkstone and pulverized material and may embrace a contribu-
tion of soils and subsoils swept up by the rock-torrent.
Large rock-falls on steep slopes obtain sufficient momentum to
carry them a considerable distance over gentle and even reversed
gradients at the mountain’s base. In the Elm rock-fall of 1881
1 James Geikie, Prehistoric Europe (London, 1881), p. 219.
2 Whitman Cross, ‘Geology of the Rico Mountains, Colorado,” U.S. Geol. Surv.,
21st Ann. Rept., Part II, p. 149.
166 STUDIES FOR STUDENTS
the landslide mass, after reaching the foot of the mountain, poured
down the level valley floor for nearly a mile, covering it over its
whole width to a depth of more than 30 feet.t. The section of a
rock-fall breccia taken normal to the cliff therefore resembles that
of the rock-glacier. No marked difference in the size of fragments
in different parts of the area covered has been recorded, so far as
the writer is aware. In the breccia formed by the landslide at
Frank, Canada, in 1903, described by McConnell and Brock,’
large rocks are said to be common everywhere.
In rock-fall breccias, fragments are of local derivation and of a
limited number of kinds of rock. No zonal arrangement is possible
in a single slide; but where repeated falls occur, the earlier bringing
down material from higher horizons on the mountain and the later
falls material from lower terranes, zonal arrangements both vertical
and horizontal may result. Thus Howe? explains the zone of
rhyolite which forms the outer rim of the Pierson Basin rock-
stream in Colorado, while the center of the mass consists of frag-
ments of andesite of a lower volcanic series.
In rock-slides the movement is gradual and repeated. The
displaced mass, therefore, is not so completely broken up as in
the rock-fall, nor does it come to cover so large an area, except
under special conditions. No type of breccia contains such large
blocks as this. While the largest blocks of the Frank rock-fall
measured 4o feet, blocks 300 feet in diameter occur in the Rico
rock-slides, and in the profounder displacements of the adja-
cent telluride quadrangle one block of more than two miles in
length has been described.4 On the north side of the Grande
Ronde River, Idaho, the Columbian lavas supply slidden blocks
half a mile in length.s The breccia of rock-slides is characterized
by the confused relations of the slipped blocks, their varying dip
Sir W. M. Conway, The Alps from End to End (Westminster, 1895), 3d ed.,
p. 246.
7R. G. McConnell and R. W. Brock, Ann. Rept. Dept. of the Interior, Canada,
1903, Part VIII.
3 Ernest Howe, ‘‘Landslides in the San Juan Mountains, Colorado,’ U.S.
Geol. Surv., Prof. Paper 67, p. 34.
4 Whitman Cross, U.S. Geol. Surv., 21st Ann. Rept., Part I, p. 146.
5]. C. Russell, U.S. Geol. Surv., Water Supply Paper 53, p. 76.
STUDIES FOR STUDENTS 167
and strike, their fissured and shattered condition, and their merging
in places into rubble breccia of small fragments due to the breaking
up of the blocks either by the force of the original slide or by that
of later movements. A diagnostic may be found in the inward dip
of the blocks, as noted by Russell.‘ Owing apparently to friction
with the surface, the sliding blocks tend to rotate backward on
axes normal to the slope and thus come to rest with a marked dip
toward the cliffs from which they were derived. On the other
hand, blocks creeping down talus maintain a dip parallel with the
slope. ‘The upper surface of both rock-fall and rock-slide breccias
is hummocky, but this feature will hardly be preserved eS
where the landslide fell into deep water.
Bajada breccia.—The rugged mountains of arid regions are
commonly fringed with wide slopes of rock-waste called bajada.
Like those of talus, the fragments of the bajada have been detached
by mechanical weathering. But unlike other subaérial breccias,
the bajada has been aggregated largely by intermittent streams.
Its fragments have been more or less water-worn. It forms an
imperfect breccia and yet is far from being a typical conglomerate
of well-rounded pebbles. The streams which build the bajada
have certain peculiarities which greatly lessen the wear on the
stones they carry. The long accumulated waste on the mountain
slopes swept down by spasmodic rains loads so heavily the tempo-
rary streams of the barrancas that they have been designated as
mud-flows.?, In the washes and on the lower unchanneled slopes
of the bajada the viscosity of the flow is further increased by
absorption of the water by the thirsty sands. In the moving mass
of the mud-flow, stones such as are carried in mountain torrents
of humid climates as the bottom load and dashed against one
another and the stream bed are here intermingled with the finer
waste held in suspension, and are thus protected from mutual abra-
sion. Hence pebbles remain imperfectly rounded even to the outer
edge of the bajada slope.
tT. C. Russell, ‘Geology of the Cascade Mountains in Northern Washington,”
U.S. Geol. Surv., 2oth Ann. Rept., Part II, p. 10.
2 For a graphic description of a mud-flow in the Himalayas see Sir W. M. Conway,
Geog. Jour., II (1893), 201.
3R. D. Oldham, Quar. Jour. Geol. Soc. London, L (1894), 460.
168 STUDIES FOR STUDENTS
Unlike all other subaérial breccias, the bajada is stratified, and
unlike that of marine and ordinary fluviatile conglomerates, the
stratification is imperfect. The bedding is often local and ill-
defined. Unsorted beds pass both vertically and _ horizontally
into those well sorted. All this follows from the nature of the
building-streams. Leaving the barrancas of the mountains and
debouching on the gentler slopes, they suffer a sudden arrest of
velocity as well as diminution of volume by the absorption of their
waters. ‘These checks are often enough to cause them to throw
down their load, fine and coarse alike, with little or no sorting.
The attitude of unequiaxed fragments of mud-flows on bajadas has
not been sufficiently observed. It may be inferred from the method
of their carriage that they will not always take the position of
repose and that even flat stones may be left at any angle.
A bajada breccia is wedge-shaped, but on a larger scale than that
of talus. Furthermore, there is a gradual decrease in size and
angularity of the material in passing from the thick to the thin end.
The matrix of the bajada consists of the finer stream-wash and
of dust and sand contributed by the wind. Beds of rubble breccia
with interstitial matrix pass into endostratic breccias of predomi-
nant matrix and sporadic fragments. A characteristic matrix is
a chemical deposit of lime carbonate by calcareous evaporating
water. As in the Tintic mining district, Utah,’ the material may
thus be cemented into compact rock in which roofs and walls
of deserted tunnels remain standing for years untimbered. The
limy matrix may whitewash the pebbles and, by filling the inter-
stices, form beds of caliche.
Dry climate conditions are indicated further by the absence
of carbonaceous deposits and by the complete oxidation of iron
compounds, the interfingering of playa clays about the outer
margin, and, under extreme aridity, the association of beds of salt
and gypsum in the centers of the bolsons. Wind-carved pebbles
may be found, and beds of the millet-seed sand of the desert. The
presence of well-rounded sand grains in a breccia, however, is a
criterion to be used with intelligent caution. Desert sand once
1G. W. Tower and G. O. Smith, U.S. Geol. Surv., roth Ann. Rept., Part III,
pp. 668-69.
STUDIES FOR STUDENTS 169
shaped retains its form indefinitely. Thus the St. Peter sandstone,
whose grains were originally ground under desert winds, was
spread by the Ordovician sea, and again supplied the material
for the Sylvania sandstone of Michigan. Either of these sand-
stones could furnish desert sand to breccias of several different
types.
Bajada breccia is of local derivation; far-traveled stones from
distant sources are not expected, yet the extent and complexity of
the feeding-mountains may give the breccia great lithological
heterogeneity. Ancient breccias are not necessarily associated
with the buried mountains or uplands which supplied their waste.
Such elevations may have been destroyed in the building of the
bajada or by later denudation.
The Permian breccias of the Midlands, England, are now com-
monly regarded as of bajada origin... They occur in rudely strati-
fied, wedge-shaped masses, some more than 200 feet thick at one
end and thinning out within four to eight miles. Beds of breccia
are interstratified with current-laid sandstone and with marl, into
which they graduate both vertically and horizontally. The frag-
ments embrace a large variety of rocks and are now considered of
local derivation. They are angular and subangular, more or less
water-worn, but are never well rounded. Half a foot is a common
measure of their size where they are largest. The matrix is cal-
careous or sandy.
Certain coarse breccias, intercalated with thin sandstone layers,
in the Esmeralda formation of Nevada have been classified as
detrital-slope breccias by Turner.2 Examples well known to
American students are the bajada breccias of the Newark forma-
tion.
Glacial breccia.—Subaérial glacier deposits are certainly to be
classified as breccia, but their characteristics are so well known and
so easily recognized, as a rule, that no description is considered
needful.
tR. D. Oldham, Quar. Jour. Geol. Soc. London, L (1894), 463-70; W. W. King,
ibid., LV (1899), 97-128; T. G. Bonney, zbid., LVIIL (1902), 185-203.
2H. W. Turner, ‘Geology of Silver Peak Quadrangle,” Bull. Geol. Soc. Am.,
XX, 245.
170 STUDIES FOR STUDENTS
Volcanic subaérial breccia.—Volcanic breccias laid in open air
include:
t. Flow breccia, in which unrounded fragments have become
incorporated in flowing lava, either from its frozen and broken
crust or from material, volcanic, residual, or of other origin, over-
ridden by the advancing stream.
2. Tuff breccia, made up of the fragmental products of explo-
sive eruptions. The matrix consists of the finer materials of the
eruption and in some instances has been washed in by mud-flows.
Or the matrix may be formed by gangue and ore stuffs deposited
in the interstices by heated waters. Tuff breccias often include
fragments of the country rock torn from the sides of the duct below
the base of the volcanic cone.
Subaqueous breccias —Breccias accumulated under water may
be divided into three general classes with the following sub-
divisions:
1. Breccias of subaérial fragmentation
a) and 6) Subaqueous talus and landslide breccia
c) Raft breccia, deposits from rafts of ice (iceberg or ice floe), trees, or
seaweed.
d) Desiccation breccia
e) Subaqueous volcanic breccia
2. Breccias whose fragmentation is the work of aqueous agencies or of agencies
working in water
a) Shoal breccia
b) Reef breccia
c) Beach breccia
d) Tide-glacier and shore-ice breccias
3. Breccia whose fragmentation is due to internal stresses
a) Glide breccia due to overload, earthquakes, deformation, undercut
Talus and landslide subaqueous breccias.—These two varieties
may be taken up together because of certain common features.
Both consist of local beds of angular fragments from near-by
sources, intercalated between younger sedimentary strata. Suit-
able topographic conditions for the formation of each are found in
the fjords and rias of a rugged coast and in mountain lakes, espe-
cially Chelans, lying in oversteepened glacier troughs. ‘The matrix
is partly of the same material as the fragments and partly of infil-
STUDIES FOR STUDENTS 7
trating sediments. It is generally interstitial, in contrast with the
preponderant matrix of raft breccias, but about the margins blocks
projected farthest may be found sporadic.
Talus and landslide breccias may be discriminated from each
other by their characteristic profiles—the even and smooth slope
talus and the hummocky surface of the landslide—by the longer
extension of the landslide from the parent cliffs, and, in the follow-
ing example, by the landslide‘tearing up the sediments of the sea-
floor over which it moves and mingling them with its own débris.
The very interesting Jurassic breccia of the Ord, on Moray
Firth, Scotland, is pretty certainly due to a landslide fallen into the
sea... The strata among which the breccias are imbedded consist
of finely laminated shales with occasional thin seams of limestone.
Hence the water in which they were laid was quiet, unvexed by
powerful waves or currents. The littoral fossils—ammonites,
corals, etc.—show that the deposits are marine and indicate the
close proximity of the shore. This is confirmed by numerous
remains of cycads, ferns, and conifers apparently drifted in by rivers.
Both fauna and flora prove the warmth of the climate and forbid
the assumption of an ice raft as the means of transportation. ‘The
breccia is contemporaneous with the Jurassic beds in which it lies,
but it is not intraformational, since its fragmental material is
derived from the Old Red Sandstone which occurs in the immediate
vicinity. The fragments vary in size from chipstone to blocks 10
feet in diameter. The majority are sharply angular, some show
signs of attrition on the edges, and not a few, especially those of
smaller size, are completely rounded. They are heaped together
in the wildest confusion. The upper surface of the breccia beds is
irregular, and the strata deposited upon it show the influence of
its projections. The breccias vary in thickness from a foot or two
to 50 feet. The matrix is fossiliferous with contemporaneous
Jurassic fossils in a more or less comminuted condition. In places
are found numerous masses of Jurassic reef-building coral torn from
their bases and heaped in all positions among the débris.
«J. W. Judd, ‘The Secondary Rocks of Scotland,” Quar. Jour. Geol. Soc. London,
XXIX (1873), 187-95; J. F. Blake, “On a Remarkable Inlier among the Jurassic
Rocks of Sutherland,” ibid., LVIII (1902), 290-310.
172 STUDIES FOR STUDENTS
All these phenomena are explained by landslides descending
from the steep slopes of Old Red Sandstone hills or mountains into
the quiet waters of a Jurassic fjord or ria and depositing on the
even-layered silts the tumultuous beds of breccia. ‘The débacle
would sweep up rounded pebbles from the beaches, tear corals
from their bases, crush shell banks in the estuary, and mingle their
débris with the fragments of the slides. That earthquakes were
the cause of these great rock-falls is suggested by the contemporary
sandstone dikes found in the district.
It may be added, if only in illustration of the diverse interpre-
tations held of breccias, that Murchison? described these breccias
as due to crush incident to the upheaval of neighboring granite.
Blake} argues the deposit of an ice foot. Huddleson, in the dis-
cussion of Blake’s paper, postulates ocean currents strong enough
to tear up masses of corals and to gather and distribute old shore
accumulation of talus, although ocean currents, even if powerful
enough to transport the immense blocks of the breccia without
wear of edges, are not so paroxysmal as to heap them in the midst of
the fine silts of quiet water. Judd‘ recognizes the cataclysmic nature
of the formation and suggests very tentatively river-floods of the
most violent character. Yet the floods of a river cannot be expected
so to maintain their energy on entering the ocean as to deposit
their bottom load in water of considerable depth and to mingle it
with detritus torn from the ocean floor. The momentum of the
rock-fall would seem to be the only force capable of the work, and
this origin is advanced by Woodward’ and by Teall® in the dis-
cussion of Blake’s paper.
Raft breccias.—In breccias of this class the fragments have been
transported in such a way as to escape wear en route. Angular
fragments of such soft rocks as shale and talcose, schists and lime-
stone, brought unworn from distant sources, prove that the car-
riage was upon the surface of the ocean, and not by wave and
current along the ocean floor. Further evidence of surface trans-
tH. B. Woodward, Quar. Jour. Geol. Soc. London, LVIII (1902), 206.
2 Transactions Geol. Soc. London, Ser. 2, Vol. II, Part II, p. 293.
3 Op. cit. 4Op. cit. 5Op. cit.
6 Quar. Jour. Geol. Soc. London, LVIII (1902), 205.
STUDIES FOR STUDENTS 173
port should be looked for in disturbances in the bedding of the
inclosing strata. LLaminae beneath the larger stones may be bent
down, and the succeeding laminae may show the influence of the
projecting blocks, thus proving that the fragments were dropped
through some depth of water. Confirmation has been found in the
position of fragments with the heavier ends downward.
Raft breccias are endostratic and the fragments are sporadic.
There may also be an irregular distribution of them—in places
a huddle of fragments where the unloading of the raft was sudden.
Some blocks may be quite too large for wave and current carriage.
Rafts capable of transporting the material of breccias are either
of ice or of vegetation. Ice rafts include both icebergs and shore
_ ice in the form of floes, or of the ice-foot.
Iceberg breccia.—Since the iceberg is detached from the tide
glacier, iceberg breccia is composed of the material of the ground
moraine. In a larger or smaller proportion the fragments prove
their derivation by their subangular form and striated faces.
A considerable lithologic variety is to be expected, since the parent
glaciers usually drain a large extent of country. Transport from
distant sources has long been looked upon as evidence of iceberg
carriage, since icebergs drift farther than other rafts.
In weighing the evidence of iceberg breccia, and of glacier
breccia as well, it is often necessary to discriminate glaciation
of pebbles from slickensides by earth movements which affect
the mass of the formation. In favor of glaciation is the incrusta-
tion of planed or striated surfaces by marine organisms, such as
serpula or shells, since these surfaces must have been produced
before the deposit of the pebbles in the breccia beds.*
Striae may be considered “‘rutsch striae”? produced after the
deposit of the breccia under the following conditions: (1) when they
occur on matrix as well as pebbles; (2) when they are found on
different planes below the surface of the pebbles; (3) when they
affect traceable planes or zones of shear; (4) when the striae of
different pebbles in the same plane run in the same direction and
11. C. Russell, “Second Expedition to Mt. St. Elias,” U.S. Geol. Surv., 13th Ann.
Repft., p. 25; W. J. Sollas and A. J. Jukes, “Included Fragments of the Cambridge
Upper Greensand,” Quar. Jour. Geol. Soc. London, XXIX (1873), 11-15.
174 STUDIES FOR STUDENTS
correspond in direction with earth movements recorded in adja-
cent strata; (5) when the number of striated pebbles in different
parts of the breccia varies directly with the amount of shear;
(6) when the striae on faulted pebbles end at the fault plane;
(7) when the striated surface is covered with films deposited from
solution. Several of these diagnostics are mentioned by Marr’
as characteristic of scored pebbles having the form of glacial
bowlders in an English breccia. Tectonic breccias often display
slickensided blocks, but they are hardly liable to be confounded
with glaciated pebbles. It may be added that the proportion of
glaciated pebbles in iceberg breccias may be exceedingly small in
comparison with that in the drift-sheets of far-traveled continental
glaciers.
Iceberg breccia, as well as any other, may be sheared after its
formation. In this case neither slickensides nor glaciation can be
used to disprove the other process. A dual origin seems to be indi-
cated in scored pebbles of some of the Permian breccias of England,
but there are students who claim that they are due to earth
movements only.
Shore-ice breccias——In arctic regions shore ice often receives
a load of angular waste, and, drifting along the coast or out to sea,
desposits it as breccia amid the sediments of the ocean bed.
The ice-foot, described by Feilden and De Rance? as having its
origin chiefly in snows drifted into water offshore, receives the
waste of the talus slopes at whose base it lies. Ice floes along shore
also obtain a load of similar débris tobogganing out from talus
slopes and falling upon the floes from sea-cliffs. Shore ice may
also carry rounded beach pebbles frozen to its base and glaciated
pebbles shaped by the grinding of ice pans on shelving shores in
storms and under the action of the tide.
The Quebec group of the Ordovician of Canada contains breccias
explained by Sir J. William Dawson as early as 1833 as due to shore
«J. E. Marr, ‘Notes on a Conglomerate near Melmerby,”’ Quar. Jour. Geol.
Soc. London, LV (1899), 11-13.
2 E.g., the Wapsipinicon breccias of Iowa, W. H. Norton, Iowa Geol. Surv., TX,
447-48.
3H. W. Feilden and C. E. de Rance, ‘‘Geology of the Arctic Coasts,” Quar.
Jour. Geol. Soc. London, XXXIV (1878), 563-606.
STUDIES FOR STUDENTS 175
ice! These breccias are very irregular in their distribution and
vary rapidly and greatly in their thickness. The fragments are of
Cambrian limestone and of the lower limestones of the Quebec
group. “The only means of explaining these conglomerates seems
to be the action of coast ice . . . . which seems to have had great
reefs of limestone, probably in the area of the Gulf of St. Lawrence,
to act upon and to remove in large slabs and bowlders, piling these
up on banks to constitute masses of conglomerate.” Walcott also
postulates floating ice in the absence of any other explanation in
accounting for bowlders in certain intraformational conglomerates,
saying: “No other explanation occurs to me that will account for
the transportation of a bowlder from the shore line and the placing
of it upon the sea-bed so as not to disturb to any marked degree
the sediments then accumulating.”
The bowlder beds of the Talchir group of India are attributed
to ice rafts by Oldham. Sporadic bowlders from distant sources
and reaching a maximum diameter of 15 feet are distributed with
extreme irregularity in distinctly stratified shales and sandstones.
Large numbers occur within limited tracts, but over many square
miles of the area they are quite absent. Where the sedimentary
matrix is laminated, the laminae bend down beneath and arch
over the included blocks. As the fragments are far too abundant
and widespread to have been carried by rafts of vegetation, floating
ice remains the only possible vehicle. ‘This inference is confirmed
by the presence in two localities of striated pebbles, although most
of the fragments are distinctly water-worn. The various phe-
nomena of the Talchir beds point to their accumulation in large
inland water bodies covered with ice in winter, to which torrential
streams led down steep valleys and to which glaciers locally
descended.
Tree-raft breccia.—Uprooted trees, drifted down to sea on river-
floods, may carry for some distance out from shore angular stones
Sir J. W. Dawson, ‘On the Eozoic and Paleozoic Rocks of the Atlantic Coast
of Canada,” Quar. Jour. Geol. Soc. London, XLIV (1888), 809-910, quoting an earlier
paper.
2C. D. Walcott, ‘‘Intraformational Conglomerates,” Bull. Geol. Soc. Am., V, 197-
3R. D. Oldham, Geology of India, 2d ed., pp. 157-60.
176 STUDIES FOR STUDENTS
of the waste mantle and of the weather-broken rocks beneath,
firmly held entangled in the meshes of their roots. ‘The fragments
may be expected ‘to be smaller than those of the ice raft. Fine
waste is absent from the breccia, since it is soon washed out of the
interlacing roots of trees when immersed in river or sea. Since
the fragments are deriyed from the zone of weathering, decom-,,
posed and etched surfaces and weather-rounded edges may be
looked for, especially in fragments made of the more soluble rocks.
Seaweed breccia.—The buoyant power of seaweeds and the
tenacity with which they adhere to rock are well known. They
are thus able to transport stones so small that they would readily
escape from the roots of floating trees. At the Orme’s Head, North
Wales, angular fragments of limestone have been found attached
to the roots of Laminaria.t’ The stones which seaweed commonly
carry are the well-worn shingle of the pebble beach. But angular —
stones may be transported by them when fragments are broken
by the battering of waves from the rocky reefs on which seaweeds
grow.
A pudding breccia occurring in one or two localities near Dublin,
Ireland, has been attributed to tree rafts by Jukes? and to carriage
by seaweed by Ball,3 although earlier observers had invoked rafts of
floating ice. The matrix is highly fossiliferous, encrinital, car-
boniferous limestone. ‘The fragments, sharp-edged, sporadic, small,
are of granite and metamorphic rock outcropping in the neighbor-
hood. ‘The small size of the fragments lends some weight to sea-
weed as the transporting agent.
Desiccation breccia.A—Surface layers of unconsolidated fine-
grained sediments, such as clay or limy mud, when exposed to the
air, dry, shrink, and sun-crack. ‘The angular blocks of this mosaic
may again be covered with water and imbedded in the sediments
which it throws down. The conditions for desiccation breccia
are afforded where there are long intervals between periodic
tC. E. de Rance, Quar. Jour. Geol. Soc. London, XLIV (1888), 374.
2 Jukes, Manual of Geology (1886), p. 298.
3V. Ball, Quar. Jour. Geol. Soc. London, XLIV (1888), 371-74.
4 Desiccation conglomerates is a term proposed by J. E. Hyde, Amer. Jour. Sci.,
4th ser., XXV, 400 f.
STUDIES FOR STUDENTS 177
floodings, as in the shallow lakes of arid basins and those of river
flood-plains and in lagoons cut off from sea except at highest tide
or greatest storms. Desiccation breccia may be only as thick as
the sun-cracked layer. Where the dried blocks of the mosaic or
pieces of their upturned edges are assembled by the waves, the
fragments may be irregularly piled in rubble and should show some
wear. ‘The matrix differs little from the fragments, and the breccia
is endostratic. A special variety is playa breccia. The cracks °
of the sun-baked clay of the dried-up lake bed may be filled with
desert sand, and this accumulates also beneath the curled-up edges
of the cakes. Desiccation breccias have been described from the
Algonkian of Idaho by Ransome and Calkins," and designated as
‘“‘mud breccias.” The angular or slightly rounded fragments are
of argillite and are imbedded in a somewhat coarser-grained and
more arenaceous matrix. Sun-cracks are found in direct connec-
tion, and the angular fragments are supposed to be broken off from
the edges of flakes of mud curled up by drying in the sun.
Volcanic subaqueous breccia.—Volcanic breccia deposited under
water may be distinguished from that laid on land by the sediments
on which it rests and by the bedding of the tuff. Subaqueous
tuff breccias, as remarked by Leith,? are distinguished only with
very great difficulty from water-laid clastics resulting from the
erosion of volcanic rocks.
Shoal breccia.—In this class of submarine breccias, and in reef
and beach breccias as well, disruption and assemblage both are
caused by waves and tides. The normal action of these agents is
to round and sort the coarser stuff they handle and to deposit it in
well-defined conglomerates. It is only under exceptional condi-
tions that they can assemble peas of fragments so little worn as
to constitute a breccia.
Shoal breccia is formed by the action of waves and tides on
shoals due to diastrophic movements or to general aggradation.
In reef breccia, on the other hand, there is proof that the shallows
permitting wave-pluck are due to local upbuilding of the sea-floor.
tF, L. Ransome and F. G. Calkins, ‘“‘Geology and Ore Deposits of the Coeur
d’Alene District, Idaho,” U.S. Geol. Surv., Prof. Paper 62, p. 31.
2C. K. Leith, Structural Geology (New York, 1913), p. 66.
178 STUDIES FOR STUDENTS
Shoal breccias are commonly of limestone. Calcareous sediments
rapidly harden by cementation, and may be broken into breccia by
waves which under identical conditions merely redistribute the
grains and particles of unindurated sands and clays. On shoals
of calcareous sediments lying partly above and partly below the
plane of effective wave-erosion, waves and tides tear up the
cemented beds of the elevations and deposit fragments in the
hollows safe from further wear.
In the case of the extensive sheets of brecciated limestone of
the Galena and the Niagara formations of Wisconsin, Chamberlin’
has suggested that the tide may have played an important rdle.
Under a large tidal oscillation storm waves may be brought at low
tide within reach of the surface of shoals which, except at this
brief interval, remains below wave-base. Fragments torn at low
tide by storm waves, and the finer waste stirred into suspension,
thus have time to settle back together as fragments and matrix
of a breccia, and, it may be added, to be further protected by a
cover of other sediments before a low tide again coincides with a
heavy storm and the process is repeated.
Strong tides working on shoals are postulated by Lane?’
in explaining the limestone breccias of the Salina and Lower
Helderberg of Ohio. ‘The prevalence of ripple-marks, mud-cracks,
brecciated and conglomeratic layers, leads to the inference of a great
flat which seems to have been just awash. “If we imagine tides
like those of the Bay of Fundy rushing over this flat, producing
this breccia and conglomerate . . .. we have the conditions
of the Helderberg on Monroe deposits.”
In explaining the foundations of Mississippian reefs in York-
shire, England, Tiddeman? infers local deformations which here and
there brought the sea-floor above wave-base. As a result, shoals
were formed of wave-plucked angular and more or less rounded
fragments on which colonizing corals and mollusks reared their
reefs. In a similar way certain Algonkian breccias of Idaho are
explained by Ransome and Calkins.4 “It is supposed that in the
tT. C. Chamberlin, Geology of Wisconsin, I (1883), 168-69.
2 A.C. Lane, Geol. Surv. of Michigan, V (1895), Part II, p. 27.
3 R. H. Tiddeman, cited by Marr, Quar. Jour. Geol. Soc. London, LV (1899), 330.
4 Op. cit., p. 38.
STUDIES FOR STUDENTS 179
vast mud flats upon which the St. Regis beds were being laid down,
the surface was raised up into a low dome of small extent upon which
the soft strata were exposed to wave action and fragments were
broken away and incorporated with the soft siliceous mud that
was then accumulating in the surrounding waters.”
Reef breccia.—Reefs with brecciform structures may be built
by corals, calcareous algae, and molluscous shells. Coral breccia,
the variety most common and most closely studied, is produced in
several different ways. As a coral reef is built up toward low-tide
level, the interspaces between the masses of growing coral are filled
with broken fragments of coral branches and the finer waste of the
reef. The coral framework is brittle and is further weakened by
boring worms and mollusks; hence the accretion of broken branches
goes on below the zone of wave-wear, and the fragments remain
angular. After wave-base is reached, accretion proceeds still more
rapidly, and now the other rim of the reef, the belt of its most active
growth, acts as breakwater and protects the inner portions of the
coral fields from the wear of the surf. Thus is formed reef-rock
breccia or coralline rag, a well-cemented limestone in which masses
of coral retain the attitude and position of growth, and to which
the varied animal and vegetal life of the reef contributes.
In this reef-rock waves cut the channeled and cavernous rock-
bench. The fragments plucked from the bench are swept inland
by heavy storms over and beyond the beach of coral sand, and
cover large tracts with lichen-blackened fragments, angular to
such a degree that both Dana‘ and Sollas? have compared them
to the rough clinkers of lava which strew the slopes of Mauna Loa
and of Etna. Intermixed is wave-worn and wind-blown coral
sand, which acts as a matrix, cementing the breccia of the island
rock. By slow subsidence these deposits may be carried beneath
the surface of the sea. The upgrowth of the rim of the reef mean-
while protects them from being worked over by the waves and thus
the brecciated stucture is preserved.
A third variety of coral breccia accumulates at the foot of the
steep outer face of the reef, where angular fragments torn by waves
tJ. P. Dana, Corals and Coral Islands (New York, 1879), p. 178.
2W. J. Sollas, Age of the Earth (London, 1905), chapter on Funafuti, p. 108.
180 STUDIES FOR STUDENTS
from the growing corals of the rim come to rest below wave-base.
Such a breccia, at the foot of reefs of Mississippian limestones in
Great Britian, has been described by Tiddeman.?
The characteristics of coral breccia may be enumerated:
1. Like all wave and tide breccias, coral breccia is either a
rubble or a pudding breccia. Crackle and mosiac breccias are not
to be expected.
2. The matrix consists of the fine detritus of the reef.
3. Both matrix and fragments are singularly devoid of siliceous
and argillaceous impurities. An exception occurs in reefs which
receive more or less waste from an adjacent land.
4. Reef-rock breccia may show little or no trace of bedding.
In the core of the deep boring of the Funafuti atoll, which passes
through this rock to a depth of 1,114 feet, the only stratification
found was that due to such irregular accumulation of detrital
material as occurs between and around the corals.2,_ The numerous
Silurian reefs of Wisconsin and Iowa show little or no trace of
bedding from top to bottom, while areas occur within them of con-
glomeratic or brecciated structure. In the case of the wave-driven
fragments of the island rock some sorting and bedding with low dips
are to be expected, and the talus formed below the reef probably
shows rude layers dipping outward at the angle of repose.
5. The fragments of coral breccia show varying amounts of
wear. Least worn are fragments of reef rock accumulated below
wave-base. The island rock necessarily approaches a conglomerate
in the rounding of its constitutent masses. How short a time and
distance are needed to destroy the angularity of fragments is seen
in a photograph and description by Kent? of the result of a single
tropical hurricane of a few hours’ duration in 1884. A fringing
reef was wrecked and its fragments, swept inland, were piled above
reach of the highest tides. Massive head-corals were torn up and
rolled together like the small pebbles of the beach and ground down
to subspherical symmetry.
1R.H. Tiddeman, Rept. British Soc. (Newcastle-on-Tyne), p. 602.
2 Judd, quoted by Sollas, op. czi., p. 128.
3T. C. Chamberlin, Geology of Wisconsin, I, 184; W. H. Norton, Zowa Geol.
Surv., UX, 424; XI, 307.
4 Saville Kent, Greai Barrier Reef of Australia, pp. 50-52.
STUDIES FOR STUDENTS 181
6. Coral breccia is intimately associated with stratified deposits
of coral mud and sand and pebbles. ‘The reef contains stretches
of barren sand within the outer rim. The island is bordered by a
beach of sand and shows extensive tracts of sand in the interior.
Soundings disclose belts of sand with which the talus of the reef
must interfinger. Fine-grained limestones are forming in the
lagoon and in deeper offshore waters.
7. Fragments may themselves be brecciform. Complex brec-
ciation occurs especially where fragments of the reef-rock breccia
are carried inland to form the island rock.
8. The chief diagnostic of an ancient coral breccia is the presence
of the reef proved to be the work of corals by its fossils. Thus the
classification of the breccia of the St. Louis formation of south-
eastern Iowa and adjacent parts of Missouri and Illinois as a coral
breccia is held untenable by Bain" because of the absence of
reef-building corals. On the other hand, brecciated structures
connected with coral reefs are not necessarily of coral origin.
Associated with the Silurian reefs of Iowa are local breccias un-
questionably due to the later deformation of beds of limestone
accumulated upon the flanks of the coral mounds.
Beach breccia#—On beaches where wave-action is inefficient
and angular blocks are supplied as from sea-cliffs, a deposit of
fragments so little worn as to be classed as breccia according to
prevailing usage may result under conditions of rapid submergence.
In all cases, however, more or less wave-wear will be found upon
the fragments, and the deposits, like other subaqueous deposits of
both angular and rounded material, should perhaps be termed
a breccia-conglomerate.
The St. Louis breccia of southeastern Iowa, classified by Gordon?
as reef breccia, is considered by Bain’ a shore formation in which
blocks of limestone up to 4 feet in diameter were torn from their
beds and buried in sands apparently at the foot of a series of cliffs.
Savage’ also finds evidence of vigorous wave-action and a close
1H. F. Bain, Jowa Geol. Surv., V, 150.
2C. H. Gordon, ‘On the Brecciated Character of the St. Louis Limestone,’
Am. Naturalist, XXIV (1890), 305-13; Jour. Geol., III, 289 f.
3H. F. Bain, Jowa Geol. Surv., V, 150.
4T. E. Savage, Iowa Geol. Surv., XII, 263 f.
182 STUDIES FOR STUDENTS
proximity of the shore. ‘The subaqueous origin of the breccia is
confirmed by Van Tuyl,’ but only inpart. The first period of dis-
turbance was one in which, under violent wave-action, mounds of
shoal breccia were produced. ‘The major disturbances, however,
are of later date and gave rise to tectonic breccias associated with
mashing, folds, and overthrust faults.
Tide-glacier breccia.—Tide glaciers, laying their loads on sea-
bottom, give rise to breccias which prove their parentage by faceted
and scored pebbles of a considerable variety of rocks and by sub-
jacent disturbed sedimentary deposits or glacier pavements
- where the ice has overridden the sea-floor. Associated stratified
beds with littoral fossils prove the breccia submarine. Icebergs
detached from the glacier front extend the formation seaward in
an ice-raft breccia, with a lessening proportion of morainal stuff.
The Chaix Hills, described by Russell,? are carved from an
uptilted block 4,000 or 5,000 feet thick, composed of morainal
material and sea-clays. The fragments of this breccia are sporadic
throughout the terrane from base to summit. They are both
angular and rounded and reach a diameter of some 8 feet. Litho-
logically they are as various as are the bowlders of the moraines
of the living glaciers of the encircling mountains. Sea-shells of
living species are numerous in the finer portions, which are largely
made of glacier silts. .
Glide breccias —The sediments of the sea-floor are subject to
slow and rapid gravitational movements, comprehensively termed
glides, which deform, shatter, and brecciate the involved strata.
Glides may be expected to affect the steeper slopes, such as the
sides of submarine channels, the front of deltas, and the edges of
continental shelves. They are known to have taken place on
slopes as low as about three degrees. The mobility of marine
deposits is increased by permanent saturation and frequently by
lack of cementation.
Subaqueous. glides may be classified according to their chief
precipitory causes as overload glides, earthquake glides, and deforma-
tion glides.
tF. M. Van Tuyl, ‘‘Brecciation Effects in the St. Louis Limestone,” Bull. Geol.
Soc. Am., XXVII, 122-24.
21. C. Russell, “Second Expedition to Mt. St. Elias,’ U.S. Geol. Surv., 13th
Ann. Rept., pp. 24-26.
STUDIES FOR STUDENTS 183
Overload glides: On land, gravitational movements occur on
slopes due chiefly to erosion; on the sea-floor unstable equilibrium
must often result from aggradation. Sediments are unequally
spread owing to set of current and distance from sources of supply.
Deltas and banks are thus built up, until along their edges overload
gives rise to facial shear and glide.
Earthquake glides: The chief geologic effect of earthquakes
on land is to precipitate movements both of the waste mantle and
of the solid rock beneath. Alluvium on valley floors lurches toward
the thalweg, the waste on hillsides, slumps and avalanches, and even
solid rock may be intimately shattered and shaken down in land-
slides of the first magnitude.* The effects of earthquakes on
marine deposits must be of similar nature and proportionally great.
Evidence collected by Milne? proves conclusively the fact of sub-
aqueous glides and their close connection in a number of instances
with earthquakes. Since the continental delta throughout geologic
time has been the zone, not only of sedimentation, but also of
great diastrophic movements of which earthquakes are an expres-
sion, it may be assumed that earthquakes have been a not uncom-
mon cause of glides in geologic history. Yet no instance is known
to the writer in which a glide breccia has been assigned to this
precipitory cause. The most direct evidence pointing to such an
origin is to be found in contemporaneous associated faults or sand-
stone dikes. Since earthquakes recur in the same area for long
periods of time, earthquake-glide breccias may recur at successive
‘horizons in a formation on in a sequence of formations.
Deformation glides: There is some reason to believe that within
the continental delta deformation may so accent the slope that
glides of unindurated sediments result. In explaining the Devo-
nian breccia of Iowa, McGee? offered as ‘‘a useful even though a
far-fetched hypothesis”? that of an elevation at the close of the
Devonian by which the declivity was increased, a consequent slight
«The Califormia Earthquake of April 18, 1906, 1, Part II, pp. 384 f. (Carnegie
Institute, 1908); Darwin, Voyage of a Naturalist (London, 1891), p. 220; Whymper,
Travels amongst the Great Andes, IV (London, 1892), 260; Whitman Cross, “‘ Geology
of the Rico Mountains, Colorado,” U.S. Geol. Surv., 21st Ann. Rept., Part II, p. 149.
2 John Milne, ‘‘Suboceanic Changes,” Geog. Jour., X, 129-46, 259-85.
3 W. J. McGee, “Pleistocene History of Northeastern Iowa,” U.S. Geol. Surv.,
tith Ann. Reft., p. 323.
154 STUDIES FOR STUDENTS
settling seaward of the fresh-formed Devonian sediments upon the
sloping flanks of the Island of Wisconsin, and a slipping of the
strata upon one another causing crumpling, buckling, and breccia-
tion. ‘The same cause is assigned by Hershey’ for a very local
breccia near Galena, Missouri. Minor causes of subaqueous glides
are erosion and undercut of submarine banks by springs and
currents.
The characteristics of glide breccias are due to the deformation,
to the shear and crush of the gravitative movement, and not to its
precipitory cause. Hence all the varieties mentioned are alike in
structure and have a close resemblance to endolithic breccias
caused by deformation. j
The fragments are contributed by any layers hard enough to
suffer fracture. ‘They may be sharply angular or somewhat worn
by mutual attrition. ‘They may be apposed in crackle and mosaic
breccias, or disposed in rubble, according to the amount of move-
ment. Fragments may show by their relative positions the initial
attitude of the layer before fragmentation. ‘These breccias are
likely to graduate into folded structures with parallel and thick-
ened axes and common dips, and the fragments of beds bent before
breaking may show flexures and contorted laminae. A zonal
arrangement is to be looked for where strata differing lithologically
and of a considerable thickness are involved.
The matrix is supplied by the least indurated or most readily
comminuted beds, especially by the surface sediments as yet uncon-
solidated, and by any bottom layer which by its plasticity deter-
mines the base plane of the glide. Thus shales furnish matrix to
fragments broken from brittle limestones. Matrices evidently
pasty and fragments somewhat plastic at time of brecciation point
to subaqueous brecciation either by glide or by wave-action, and
the former alternative is to be chosen when there are proofs of
folding before fracture. This test applies only when brecciation
in the zone of flow and fracture is precluded. The relative amount
of matrix and fragments is determined by the volumes of strong
rock and weak rock involved and by the violence of the movement.
Even a breccia of sporadic fragments may result.
10. H. Hershey, ‘‘A Devonian Limestone Breccia in Southwestern Missouri,’
Science, N.S., I, 676-78.
STUDIES FOR STUDENTS 185
Overlying beds are undeformed or share only in later deforma-
tions affecting the entire body of the strata. Their contact is
accordant where the breccia has been reworked and leveled by
wave-action. In this case they contain at bottom fragments of the
broken beds, either rounded to a conglomerate or partially angular
and forming a pudding or rubble breccia. If the slidden mass is
reassembled below wave-base the contact is discordant and super-
incumbent beds are free from fragments. ‘The breccia may thus
be either endostratic or have the hummocky upper surface of a
landslide. Glide breccias may graduate laterally as well as verti-
cally into sedimentary beds. They rest on undisturbed strata of
earlier date. Endostratic breccias passing within the limits of the
same stratum into folded laminae point strongly to an origin in
glide. A complete section of a glide and the associated strata
would show, according to Heim,’ the following relations: in the
area bared by the glide, (1) a reduction of the number of strata as
compared with the adjacent areas, (2) local disconformity with-
out time interval; in the area on which the glided mass came to
rest, (3) increase in the number of strata, (4) superposition of older
on younger beds, (5) displacement of facies. Glides involving
subaqueous and subaérial sediments have occurred in a number of
instances on the shores of the Swiss lakes, in Sweden, and along
the Black Sea.2 In the lower Devonian limestones of Gaspe a bed
has been described by Logan’ whose structure is evidently due to
subaqueous glide. This bed, 7 feet thick, is made up of several
thin layers of limestone and limy shale, wrinkled, contorted, and in
places brecciated, while the associated beds are free from marks
of deformation.
The edgewise position of broad, flat pebbles in evenly bedded
strata near Bellefonte, Pennsylvania; has been attributed by
Brown‘ to subaqueous glide.
Endolithic breccias.—Of breccias formed within the lithosphere
the following classes may be distinguished: (1) tectonic breccia
(dynamic, pressure, friction breccia), due to crustal movements
t Quoted by A. W. Grabau, Principles of Stratigraphy (New York, 1913), p. 660.
2 Grabau, op. cit., pp. 657 £., 779 f. |
3 Sir W. Logan, Geology of Canada (1863), pp. 391 f.
4T. C. Brown, Jour. Geol., XXI, 241-42.
186 STUDIES FOR STUDENTS
and produced by lateral or vertical pressure or by tension; (2)
expansion breccia (caused by increase of volume due to chemical
change); (3) founder breccia (ablation, solution breccia), due to
the foundering of strata, usually because of the ablation by solution
of the supporting beds.
Tectonic breccia.—Three varieties are discriminated: fault
breccia, fold breccia, and crush breccia. In the latter, brecciation
is accomplished without either faulting or folding except so far as the
rupture planes of the breccia may be considered as minute faults.
Fault breccia: Here fragmentation is due either to friction along
the fault plane or to distributive ruptures associated with the major
fault, and due to shearing stresses. In stratified rock, fault breccias
associated with both normal and reversed faults are easily recog-
nized, since the zone of brecciation crosses the planes of bedding.
Friction breccias along bedding faults are more difficult to dis-
tinguish. Here one must seek for proofs of lateral displacement
in slickensides with polish, scorings, and seams of “gouge” (clays
formed by grinding). Local breccias of this variety have been
identified by Ransome’ in the Rico Mountains of Colorado.
Complex brecciation is not uncommon, since repeated move-
ments along the fault plane shatter and drag a breccia already
formed and firmly cemented with perhaps vein stuff and ore.
Breccia zones running parallel with the main fault may show but
slight displacement. Thus in the San Francisco district of Utah
such breccia zones in brittle quartzite pass downward into mono-
clinal folds in shale.2_ The rocks on opposite sides of a fault plane
may be differently affected—granite, for example, may be sheared
and the limestone opposite brecciated. Fault breccias often
form aquifers for mineralized waters which deposit matrices of
ore and gangue. Many breccias of this class have been brought
to notice because of their economic importance.
tF. L. Ransome, ‘“‘Ore Deposits of the Rico Mountains, Colorado,” U.S. Geol.
Surv., 22d Ann. Rept., Part II, p. 297.
2B. S. Butler, ‘‘Geology and Ore Deposits of the San Francisco, etc., Districts,
Utah,” U.S. Geol. Surv., Prof. Paper 80, p. 72.
3 W. H. Emmons and F. C. Calkins, ‘Geology and Ore Deposits of the Phillips-
burg Quadrangle, Montana,” U.S. Geol. Surv., Prof. Paper 75, p. 151.
STUDIES FOR STUDENTS 187
Fold breccia: Under suitable conditions of load and stress beds
may fold by means of minor fractures. Cross and parallel fracture
planes develop, and under increasing stress the fragments are
rotated and displaced. The folded bedding is more or less com-
pletely destroyed and there results a mosaic or even a rubble breccia.
Where the rock is compressed joints or fissility ruptures are pro-
duced. A rising anticline develops radial tension cracks along the
convex surface as soon as the deformation passes the limit of
elasticity of the rock. In the experiments of Willis™ the first
fractures of the arching strata occurred at points of sharpest curva-
ture—along the axial plane at the summit of the anticline and on
radial planes of shear at its base. A basal weak stratum, com-
pelled to rise beneath a competent stratum as the latter was bent
upward, accommodated itself to the change by shear, resulting in
complete brecciation.2 Here the breccia occupied the center and
base of the anticline, while the competent upper beds remained at
first unbroken and later under increased pressure were ruptured
by a stretch thrust fault.
The degree of folding necessary for brecciation varies with the
rigidity of the stratum, with load, and with the amount and rapid-
ity of application of the stress. Even limestone and granite under
slight load yield plastically and bend to a perceptible degree when
the stress is very slowly applied. On the other hand, brittle rock
under presumably sudden stresses breaks into breccia when the
deformation has not exceeded a gentle warping.’ Under the same
strain and load different rocks fold or break according to their
elasticity. Among sedimentaries the most brittle and therefore
the most liable to brecciate are cherts, some shales, and calcilu-
tites; among metamorphic rocks, quartzites, graywackes, and rather
siliceous slates. A thin layer of chert may be seen broken into
a string of angular bits within a layer of limestone which shows no
_ *Bailey Willis, “Mechanics of Appalachian Structure,” U.S. Geol. Surv., 1 3th
Ann. Rept., Pls. 75, 76.
abide Ele o3:
3 Arthur Winslow, Am. Jour. Sci., 3d ser., XLIII, 133-34; H. F. Bain, Iowa
Geol. Surv., VIII, 378.
4 Smith and Siebenthal, U.S. Geol. Surv., Geol. Atlas, Joplin Folio, p. 9.
188 STUDIES FOR STUDENTS
other trace of yielding. In the breccias of the Wapsipinicon stage
of the Iowa Devonian, a thick, tough, crystalline-granular coquina
is normally broken into large slickensided blocks which retain
something of the flexures into which the bed was thrown, while
subjacent thinly laminated calcilutites are shattered to a jumble
of small fragments."
Shales yield plastically under load, but when near the surface
and under sudden stress they easily crush to breccia. Shales pro-
mote the fragmentation of inclosing and especially of included beds
of other rocks. Thus the Wapsipinicon breccias of Iowa embrace
the Independence shale and its associated limestones. In the zone
of fracture and flowage, alternate thick layers of brittle and of
plastic rock may produce brecciated beds, alternating with folded
layers, whose arches may be truncated by the movement of the
fragments of the rigid and brecciated beds.”
Unlike glide breccias, breccias due to folding are included
between strata which have shared the brecciating deformation
according to their competency. But since any sort of breccia may
be involved along with the associated terranes in a later deformation
further proof of origin must be looked for in the remains of initial
folded structures in the breccia. A certain continuity may be
traced from fragment to fragment, showing clearly that the frag-
ments are constituent parts of a flexed and broken layer. The
breccia may graduate into folds. Where the breccia involves beds
of different kinds of rocks, anticlines and synclines may sometimes
be traced in a zonal arrangement of the crushed material.
Crush breccias: The sheet breccias of the Joplin district,
Missouri, illustrate how terranes of brittle rock may be brecciated
by lateral pressure without any further mass deformation than that
exhibited in gentle warpings. Heavy ledges of chert have been
thoroughly and finely crushed in places and cemented by a chemical
deposit from ground-water. The fragments are of small size and
are thus in direct contrast with the residual breccias of a higher
tW.H. Norton, Jowa Geol. Surv., IV, 158-61.
2C. R. Van Hise, ‘‘ Principles of North American Pre-Cambrian Geology,” U.S.
' Geol Surv., 6th Ann. Rept., Part I, p. 681.
3 W. H. Norton, Jowa Geol. Surv., IV, 165.
STUDIES FOR STUDENTS 189
horizon in the same area. ‘The breccia is endostratic and often of
the crackle or mosaic type, with fragments rotated but slightly
in the ledge."
In all tectonic breccias the fragments are left of sharpest edge
at time of breaking, and a universal sharp angularity points strongly
to a tectonic or to other. endolithic origin. Yet the fragments may
be worn by grinding one upon another in the zone of shear and thus
become so completely rounded as to be readily mistaken for the
pebbles of a wave-laid conglomerate. Thus are produced the
“‘pseudo-conglomerates”’ of Van Hise.2 The rounding of pebbles
of a conglomerate, however, is pretty uniform for pebbles of any
given size, the smaller being better rounded than the larger. More-
over, the conglomerate graduates into finer sedimentary deposits.
The rounding of the fragments of a pseudo-conglomerate is fairly
uniform for any given portion of the brecciated zone. Tracts
where fragments are well rounded regardless of size pass into tracts
where all the fragments are angular. Dale‘ has noted that angular
pebbles of soluble rock in an insoluble matrix may be later rounded
by the dissolving action of acid ground-water. It may be added
that by solution under pressure fragments develop salients and
re-entrants, wholly different from either fracture planes or round-
ing by attrition or solution. Fragments may also show in flexed
and contorted laminae evidence of the strain to which their layers
were subjected. Such distortion phenomena imply considerable
plasticity in the layers, although it was finally exceeded by the
strain. Contorted laminae may also be seen in the fragments of
breccias originating in subaqueous glides, where the plasticity of
the sediment is due to imperfect lithification.
The form and size of the fragments of tectonic breccias so far
as due to fracture depend on the amount of stress, the closeness of
joints and bedding planes, and the natural fracture of the rock.
Thus in one of the experiments of Willis’ fault planes divided a
tSmith and Siebenthal, U.S. Geol. Surv., Geol. Atlas, Joplin Folio, p. 9.
2 Van Hise, op. cit., p. 670. 3 [bid., p. 680.
4T. N. Dale, ‘Structural Details in the Green Mountain Region,’ U.S. Geol.
Surv., 16th Ann. Rept., Part I, p. 560.
5 Op. cit., Pl. 93.
19QO STUDIES FOR STUDENTS
brecciating layer at first into rhombs bounded by two faults and
two bedding planes, and afterward, under increasing pressure, into
triangular forms bounded by two faults and one bedding plane.
From brittle, thin-layered rocks under slight stresses quadrangular
and subquadrangular small fragments are derived. ‘The sharpest
edges are found in rocks of conchoidal fracture, minutely faulted
into triangular fragments.
The matrix when contemporaneous is supplied by the material
of the weaker beds involved and by wear and tear of the stronger
beds. It graduates from powder and angular sand, the product
of attrition, to chinkstone, approaching the size of the smaller
fragments. When the interstices are left at time of brecciation
more or less unfilled, as is likely to be the case, a veinstone matrix
of travertine, of jasperoid, or of iron compounds is often deposited
later by ground-water. Such a matrix weighs against any ori-
gin, e.g., subaqueous glide, which is not likely to leave unfilled
interstices. The significance of both veinstone and attrition
matrices lies also in the proof they offer that sedimentary deposits
had no access to the zone of brecciation.
The material of tectonic breccias, with the exception of the
matrix of chemical deposit, derives from the geologic formations
of the beds involved. Fragments of beds belonging stratigraphi-
cally below the base of the breccia cannot be included in it. On the
other hand, in subaqueous or subaérial breccias fragments deriving,
for example, from cliffs of Archean rock may be deposited as Devo-
nian breccia on Devonian lands or in Devonian seas. But while in
subaérial and subaqueous breccias the matrix is never older than
the fragments, in tectonic breccias an older and weaker terrane
may supply the attrition matrix, in which the fragments of a younger
superjacent stratum are imbedded. Where beds of different
rocks are involved zonal arrangement is sometimes traceable,
which at once excludes the breccia from many varieties of sub-
aqueous and subaérial origin. A tectonic breccia does not rest,
like subaérial breccias, upon an erosion surface. It cannot graduate
upward into strata inclosing sporadic fragments. An important
diagnostic may be found in undisturbed areas, perhaps of very
large size, which have transmitted the strain instead of yielding
to it by fragmentation.
STUDIES FOR STUDENTS IgI
Expansion breccia.—Fragmentation may be caused by increase
of bulk of the brecciating rock or of associated layers which trans-
mit the pressure to it. Expansion may be caused by recrystal-
lization or by hydration. Expansion breccias graduate into folded
structures, but the folds show quaquaversal deformation, the
enterolithic structure of Grabau,' a term translating the “‘ Gekrése”’
of Koken, and this intestinal coiling may serve to distinguish them
and the associated breccias from folds and breccias due to lateral
pressure.
Founder breccia.—Where beds of soluble rocks have been in
part or whole removed by the chemical action of ground-water,
founder breccias of the superincumbent beds are produced on a
scale commensurate with the extent of the ablation. All ter-
ranes between the dissolving foundation and the surface of the
ground or some competent superior stratum are affected by the
process. Characteristic features are abnormal dips, sag folds
without parallelism of axes, areas of crushed rock alternating with
horsts where the strata are undisturbed. ‘The matrix may be of
crushed material of the same terrane as the fragments. In this
case it is likely to be small in amount and insufficient to cement the
breccia firmly, for the attrition of the fragments in founder is
probably much less than in tectonic brecciation. The matrix may
be a later chemical deposit.
The fragments show only the small wear due to mutual attrition.
Like other endolithic breccias, a founder breccia can carry no
water-worn pebbles. It is conceived that founder breccias of thick
extensive beds include larger blocks than any other type of breccia
excepting that due to landslide. Certainly they may be far too
great for detachment by waves or by mechanical weathering and
for fragmentation under lateral pressure. Horsts may be difficult
to discriminate from the undisturbed areas of tectonic breccias.
Thin-shale founder breccias have been described by Ransome?
in the blankets of several mines in the Rico district, Colorado.
The blanket of the Enterprise mine, for example, is an uncon-
solidated breccia of shale from 2 to 20 feet thick, resting on a thin
t Principles of Stratigraphy, p. 757.
2F, L. Ransome, ‘‘Ore Deposits of the Rico Mountains, Colorado,” U.S. Geol.
Surv., 22d Ann. Rept., Part II, p. 273 f.
192 STUDIES FOR STUDENTS
bed of light-gray earth, shown by chemical analysis to be a residuum
of gypsum.
The Monroe breccia of Michigan is considered by Hindshaw'
to be of this type, although Lane? has classified it as a shoal breccia
and Grabau’ as a rock-stream or rock-glacier breccia, at least in
its outcrops at Mackinac Island and the vicinity. As observed
by the writer, the breccia at this locality is a rubble, entirely with-
out bedding, of angular fragments set at all angles and varying in
size from gravel up to blocks 25 and 30 feet inlength. Fragments
-and matrix are of soft buff magnesian limestone, and the latter
is usually interstitial. Occasional seams of calcite and celestite
occur. In certain areas a zonal arrangement of material is seen
in the prevalence of chert, showing that the material of the breccia
has not been intimately mingled as in subaérial breccias whose
fragmentation is due to weathering. A still more conclusive negative
to such an origin is given in areas of rock quite undisturbed, such
as a block extending eastward along the cliffs for upward of 200
feet from the eastern border of the park at Mackinac. This
observation is confirmed by Rominger,! who states that “frequently
large rock-masses composed of a series of successive ledges which
have retained their original position to each other are scattered
through the breccia.” The wide distribution of the breccia in
Michigan and Ohio precludes any local origin. The size of the
fragments and the absence of water-wear are considered incon-
sistent with a subaqueous origin. Along the eastern shore of the
island, however, numerous talus blocks show endostratic brec-
ciation: Laminae flexed and faulted and partially brecciated are
imbedded in a matrix of the same material and maintain more or
less of their original parallelism with each other and with the bedding
of the stratum. In the undisturbed block at the east of the park
at Mackinac sporadic fragments occur toward the base of massive
beds. These phenomena imply disturbed sedimentation, or sub-
t Michigan Geol. and Biol. Surv. Publ. 14, Geol. Ser. 11, pp. 206-7.
2A. C. Lane, Geol. Surv. Michigan, V, Part II, p. 27.
3A. W. Grabau, Science, N.S., XXV, 295-96; Principles of Stratigraphy, pp.
547-48.
4 Carl Rominger, Geol. Surv. Michigan, I, Part VI, p. 27.
STUDIES FOR STUDENTS 193
aqueous glides, at the time of the deposit of the limestone, but the
main brecciation is taken to be of later date.
There is also seen an association with folds which suggests either
a tectonic or a founder breccia. The underlying red and blue
shales seen in the rock bench about the island rise in places as
anticlines in the sea-cliffs. On the cliffs east of the fort at Mack-
inac, beneath a cornice of about 15 feet of horizontal massive beds,
appears a zone, g feet thick, of thin-bedded limestone, thoroughly
crackled, with unhealed seams, shattered, and in places with the
fragments rotated and displaced, but with the bedding still trace-
able to a large extent. In places low folds can be made out with
a height of 8 inches and width of some 4 feet, a deformation appar-
ently adequate to brecciate this brittle rock.
The theory of founder presupposes the removal by solution of
beds of rock salt and gypsum underlying the red and blue shales
on which the limestone rests. ‘These shales are not commingled
with the breccia of the limestone. If the breccia is due to founder,
it must be concluded that the soft clay shale yielded plastically to
unsupported pressure of overlying beds without mixing with the
fragments into which these beds were broken. In places small
chimneys of breccia penetrate the shales beneath, and the ledges of
breccia which rise in reefs above the country rock of shale in the
low ground of the north of the island may have a like relation.
The grounds on which Hindshaw’s theory of founder is sup-
ported are as follows:' The Monroe formation rests on the Salina,
which includes beds of salt, in places 800 feet in thickness, and of
anhydrite partially changed to gypsum. The Monroe itself con-
tains beds of anhydrite. In the Monroe beds and in the over-
lying Dundee limestone, which is also involved in the brecciation,
the circulation of ground-water is exceptionally active. Dis-
cordant and abnormal dips occur, accountable for by slumping due
to solution of the Salina beds about their outer edges.
Cavern breccia is a local variety of founder breccia. Detached
masses fall from the roof and sides of a cave and accumulate on the
floor as breccia, which may come to fill nearly the entire space.
Cavern breccias may sometimes be recognized by their shape as
1 Op. cit., pp. 206-7.
194 STUDIES FOR STUDENTS
casts of the irregular chambers and chimneys of caverns whose
walls remain as molds. Cave earth, bone breccia, incorporated
fragments of stalactites, and stalagmitic crusts may certify the
origin. The matrix is either of travertine or of limestone sand
and cave earth.
Campbell has described a cavern breccia exposed in the walls
of a canyon near Fort Stanton, New Mexico, and suggests that the
frequent repetition of the process might result in complete brec-
ciation of certain beds more soluble than the rocks above and
below.
Cavern breccias are common in the lead and zinc mining regions
of the Upper Mississippi Valley. Fragments of sheets of ore mingle
with the dolomite débris. ‘The matrix may be metalliferous, giving
rise to sprangle ores. Slight foundering of the strata above caverns
has produced crackle breccias whose fissures are healed with
Smithsonite.?
Pseudo-brecciation.—This term is used by Wallace’ to designate
irregular mottlings due to partial dolomitization. Various other
causes produce irregular mottlings, but such can hardly be mis-
taken for brecciation structures.
1M. R. Campbell, “Origin of Limestone Breccias,” Science, N.S., X XVII, 348.
2 Whitney, Geology of Iowa, I, 448; A. C. Leonard, Iowa Geol Surv., VI, 11 £.;
S. Calvin and H. F. Bain, zb7zd., X, 480 f.
3R. C. Wallace, Jour. Geol., X XI, 420-21.
REVIEWS
Stratigraphy of the Pennsylvanian Series in Missouri. By HENRY
Hinps and F. C. GREENE. Missouri Bureau of Geol. and
Mines, Vol. XIII, 1915. Pp. 407, figs. 5, pls. 32, map 1.
This volume treats chiefly of the stratigraphy, paleontology, and
lithology of the barren formations of the Pennsylvanian series in this
state, thus supplementing Vol. XI, which deals primarily with the coal
deposits.
The subdivisions now recognized in the series are given below:
Group Formation
Wabaunsee
Shawnee
Missouri Douglas
Lansing
Kansas City
Pleasonton (unconformity in)
Des Moines Henrietta
Cherokee
The formations in this classification are differentiated into 30 mem-
bers which might be called formations by those who give the Pennsyl-
vanian the rank of a system. The system is markedly unconformable
on the beds beneath, and its upper members are the youngest consoli-
dated rocks in the state.
The Des Moines and Missouri groups were differentiated originally
with the belief that the latter contained much greater quantities of
limestone. The only basis now recognized is a rather well-marked
faunal break. There are rather strong indications of a widespread
unconformity in the Pleasonton, and this faunal break may be related
to it. If this is found to be true the plane between the two groups will
be drawn at the unconformity. There has been much confusion in the
nomenclature of the Pennsylvanian of this area owing to duplication
of names and miscorrelations. That considerable progress has been
195
196 REVIEWS
made in correcting this situation is indicated by the fact that the classi-
fication of this report has been approved by the United States Geological
Survey and is now official for the state surveys of Kansas, Iowa, and
Nebraska with one minor exception in each of the last two states.
Lithologically, the series is made up chiefly of shales alternating
with limestones. Sandstones, clay, and coal are found in lesser quanti-
ties. Most of the lithologic units are quite persistent laterally, but
a few are notably lenticular. The broader features of the present
structure have resulted from two periods of folding since the close of the
Pennsylvanian. The first of these developed monoclinal dips to the
west and northwest and the second formed a number of narrow anticlines
with associated synclines. The axes of these folds are markedly parallel
and trend northwest-southeast throughout the state. These are shown
on a structure contour map drawn on the basis of rather meager data.
Invertebrate paleontology is the subject of an exhaustive chapter
by Dr. Girty. More than 250 collections containing more than 350
species form the basis of his report. The species are listed by localities
and by zones for each formation, and the valuable data of these lists is
made available more readily by a composite table showing the range of
each species. Descriptions and illustrations are given of a number of
species that are new or have been called into question. Paleobotany
is discussed in a short chapter by David White.
“Some progress has been made in correlating the Missouri series with
eastern areas. Paleontological evidence indicates that the lower part
of the Cherokee is of Pottsville age and the upper part is basal Allegheny.
It is suggested on the basis of incomplete collections that Allegheny
time extends to the unconformity in the LEN essen teh and that Conemaugh
time ends well up in the Shawnee.
The writers of this valuable report did not fail to include a chapter
of bibliography which includes all the important publications consulted
in its preparation.
W. B. W.
The Red Iron Ores of East Tennessee. By ERNEST F. BURCHARD.
Bull.’ Penn. Geol: ‘Survey No: 16, 1913.) Pp.)072,< pls: 7
(including 5 maps), figs. 30 (including 6 maps).
The purpose of this report is to describe and explain the origin of
the red iron ores as they occur in the Cumberland Plateau and the
Great Valley in east Tennessee.
REVIEWS 107
It is not possible to give a generalized section of the strata in east
Tennessee because of the local variations in the sequence. The ores are
contained chiefly in the “Rockwood” formation (Silurian). They are
found also in the Tellico sandstone (Ordovician) and to a very minor
extent in the Grainger shale (Devonian and Mississippian). Two wide-
spread formations, the Knox dolomite (Cambrian and Ordovician) and
the Chickamauga limestone (Ordovician), occur below the Tellico and
“Redwood.’”’ The Chattanooga shale (Devonian) and the Newman
limestone (Mississippian) which lie above the “Rockwood” are impor-
tant, the former as a reference horizon for the ‘‘ Rockwood” ore and the
latter as a source of the limestone for fluxing material in the iron industry.
In general the beds of the Cumberland Plateau are nearly horizontal,
while those of the Great Valley are tilted, folded, and faulted.
The term iron ore as used in this report includes that which runs
20 or more per cent metallic iron. The red ores consist essentially of
hematite; the impurities are calcium carbonate, silica, alumina, magne-
sium carbonate, sulphur, phosphorus, and manganese.
The Tellico ore varies from a ferruginous sandstone to lenses of com-
pact rich ore. The deposits near Riceville, near Sweetwater, and east
of Knoxville (here the ore is dominantly limonite) are described; the
two last mentioned are residual deposits.
The ‘‘Rockwood” formation is composed of lenses of sandstone,
shale, limestone, and hematite; its thickness varies from a few feet to
over 1,000 feet. The ore beds are mainly in the upper 60-200 feet.
The ore is “‘a mixture of fossil fragments and flaxseed-shaped grains.”
The soft ore (due to the leaching of calcium carbonite from hard ore)
carries 40 to 58 per cent metallic iron, while the hard ore runs from 25
to 45 per cent metallic iron. It is believed “that the ‘Rockwood’ iron
ore was formed by the deposition in a body of water of sediments con-
taining iron, together with calcium carbonate, silica, alumina, and other
minerals in minor proportions.” Later much of the calcium carbonate
of the fossils was replaced by iron oxide; this may have occurred even
before the consolidation of the strata and ‘‘it probably involved only the
original sediments.’ The “Rockwood” ore outcrops more or less con-
tinuously along the base of the Cumberland escarpment and in the
Tennessee Valley; the total linear exposure, if only a single seam is
taken into account, is 245 miles, of which 60 miles is workable hard ore.
Central east Tennessee is the most productive area in the state.
Underground (slope and adit) mining is carried on almost exclusively.
‘Notes on the Iron Industry”’ conclude the report.
Vi Ond.
198 REVIEWS
Geology and Coal Resources of North Park, Colorado. By A. L.
BEEKLY. “US. (Geol! Survey,.) Bulk S06; rors. Epa:
pls: in2: .
North Park is described as a synclinal basin limited on the east and
west by anticlinal mountain ranges, on the south by a high ridge com-
posed of Tertiary extrusives, and on the north by an area of pre-Cambrian
crystalline rocks faulted up into contact with the Paleozoic and later
sediments of the Park. The latter comprise two sharply distinct groups:
the lower, which rests upon pre-Cambrian crystallines, begins with a
few feet of limestone of doubtful age, at the base of the Red Beds, and
ends with the Pierre shale—7,400 feet of beds in all, apparently con-
formable throughout; the upper group comprises some 5,000 feet of
uppermost Cretaceous or Eocene strata included in the Coalmont for-
mation. The folding of the region took place in part prior to the
deposition of the Coalmont and in part later. Moderately extensive
faulting accompanied or followed the later folding, and therefore
affected the Coalmont in common with the older rocks. Lying upon
the Coalmont in uncertain relationship is the North Park formation,
a 600-foot series of clastic sediments interbedded with sheets of pyro-
clastic volcanics. The Tertiary igneous rocks near the southern border
of North Park occur in both intrusive and extrusive relations. Ande-
sine basalt and volcanic agglomerate are the chief types. Granite is the
chief constituent of the pre-Cambrian complex exposed in the mountains
east, west, and north of the Park.
Coal occurring in the lower part of the Coalmont formation is the
one mineral resource of this area which is now being exploited. More
than two billion tons of sub-bituminous coal are estimated as the avail-
able reserve. The coal seams are of unusual thickness, one bed reaching
a thickness of more than 50 feet, and maintaining an average thickness
of 30 feet along a 15-mile outcrop.
C2We i:
Common Minerals and Rocks: Their Occurrence and Uses. By
R. D. Georce. Bull. Colo. State Geol. Survey No. 6, 1913.
Ppw4o0, pls.15-
The main purpose of this bulletin is to describe the commoner
minerals and rocks, and furnish the means of recognizing them and
knowing their uses.
Vi Oba
REVIEWS 199
The Iditarod-Ruby Region, Alaska. By Henry M. Eakin. Bull.
U.S. Geol. Surv. No. 578, 1914. Pp. 45, pls. 6 (including 4
maps), fig. 1.
The Iditarod-Ruby region is situated in west-central Alaska between
the headwaters of the Iditarod and the Yukon at Ruby.
The geologic succession is as follows: probable Paleozoic meta-
morphic rocks; Cretaceous sedimentary and volcanic rocks; post-
Cretaceous intrusives; Quaternary unconsolidated deposits that include
glacial material.
Conglomerates (in places several hundred feet thick), the material
of which has been derived from the underlying metamorphic rocks,
occur principally near the base of the Cretaceous beds. Some contain
bowlders up to 3 feet in diameter.
Placer gold, with a minor amount of silver, is the mineral resource
of the region. The gold has been derived chiefly from quartz veins
(which are probably genetically related to the post-Cretaceous intru-
sions) that traverse the igneous, sedimentary, and metamorphic rocks.
The value of the gold and silver produced in 1912 in the Iditarod,
Innoko, and Ruby districts was respectively $3,500,000, $250,000, and
$150,000.
In 1913 the value of the winter production in the Ruby district was
$102,200, while that of the summer production was estimated at
$750,000.
VOL.
The San Franciscan Volcanic Field, Arizona. By HENRY HOLLIs-
TER RoBINSON. Professional Paper, U.S. Geol. Survey, No.
76, 1913. Pp. 213, pls. 14 (including 2 maps), figs. 36 (includ-
ing 8 maps).
The San Franciscan volcanic field embraces an area of about 3,000
square miles in north-central Arizona.
Chap. i is devoted to the geography of the region.
Chap. ii treats chiefly of the sedimentary formations and structure.
The sequence of sedimentary rocks is as follows: the Mississippian and
Pennsylvanian Redwall limestone; the Pennsylvanian Supai formation
(“Lower Aubrey” sandstone and shale), Coconino (‘Upper Aubrey”)
sandstone, and Kaibab (‘‘Upper Aubrey’’) limestone; the Permian(?)
Moencopic formation (red to light-brown shales, with some sandstone
and calcareous layers); the Triassic ““Lithodendron formation” (basal
200 REVIEWS
conglomerate, sandstone, shales, and ‘‘marls’’) and ‘“‘ Leroux formation”
(shales, with some sandstone and calcareous beds); and Quaternary
moraines and alluvium. The thickness of this generalized section is
about 2,500 feet. An unconformity occurs between the Kaibab and the
Moencopic, between the latter and the “‘Lithodendron formation,” and
between the Triassic and later rocks. The Moencopic formation is a
fluviatile or shallow-water deposit; the Triassic beds are continental
deposits. ‘The major structural feature of the region is a very flat anti-
cline which trends N. 30° W.
Chap. iii gives detailed descriptions of the volcanoes and lava fields.
Three general periods of volcanic activity are recognized: (1) widespread
basaltic eruptions from small cones, (2) eruptions of lavas (andesites to
rhyolites) to form a few large cones, and laccolithic intrusions, (3) extru-
sion of basalt (less widespread but more cones built up than in the first
named). San Francisco Mountain, which is the principal feature of the
area, is composed of “lavas and breccias belonging to five distinct stages
of eruption.”’
“The Geologic History of the Volcanic Field and Adjacent Country”’
is given in chap. iv. The volcanic activity of the first period occurred
in the late Pliocene after the peneplanation of the region, that of the
second period took place in the early Quaternary during or after the
mature dissection of the area, and that of the third period during the latter
part of the Quaternary subsequent to broad regional uplift. There
was folding and flexing during the latter half or at the close of the Eocene.
Faulting occurred at the close of the Miocene, at the close of the Pliocene,
and during the middle or latter part of the Quaternary.
The last two chapters, v and vi, are devoted respectively to pe-
trography and petrology.
VO. T.
Transactions of the American Institute of Mining Engineers. Vol. L.
New York, 1915. Pp. 1008.
Material for this volume was presented at the Pittsburgh meeting
in October, 1914. Three topics include the major portion of the volume:
(x) iron, geology, and metallurgy; (2) coal and coke with by-products;
and (3) petroleum. The volume contains less purely geological matter
than either of the preceding volumes for the year. Fifty-two papers
and discussions, many of which are illustrated, are included.
AC DB?
REVIEWS 201
Triassic Life of the Connecticut Valley. By RiIcHARD SWANN LULL.
Connecticut Geol. and Nat. Hist. Survey, Bull. 24.
The author interprets the environment, both physiographic and
climatic, of Newark time in the Connecticut valley, and gives a full
discussion of the animal life with descriptions and illustrations of both
the fossils and the trails and footprints in these beds.
The remarkable thing about this fossil field is that actual fossils are
exceedingly scarce but trails and footprints are found in marvelous
abundance. In actual fossils the invertebrates are represented by only
two species of Unio and a single aquatic insect species. The terrestrial
vertebrate skeletons are all reptilian, consisting of only two species of
phytosaurs, two of aétosaurs, and five of theropod dinosaurs.
However, the trails and footprints indicate a much greater and more
varied fauna. Of the invertebrates, annelids, insects, spiders, scorpions,
and fresh-water crustaceans of great variety were doubtless present.
The footprints represent two, possibly three, classes of terrestrial verte-
brates—amphibia of salamandrine form and also stegocephalians;
among the reptiles, lizards, turtles, and dinosaurs, and possibly, also,
rhynchocephalians, phytosaurs, aétosaurs, and theromorpha. There
is no evidence that birds were present.
Cy HE:
The Cretaceous-Eocene Contact in the Atlantic and Gulf Coastal
Plain. By L. W. STEPHENSON. Professional Paper, U.S.
Geol. Survey, No. 90-J, ro1s5. ‘‘Shorter Contributions to
General Geology, 1914.” Pp. 155-81, pls. XI-XIX (including
2 maps), figs. 13-20 (including map).
“The Cretaceous deposits of the Atlantic and Gulf Coastal Plain are
separated from the overlying Eocene and younger formations by an
unconformity of regional extent’’; the unconformity can be traced from
New Jersey to the Rio Grande, and from there southward into Mexico.
After the Upper Cretaceous sediments were laid down, the sea with-
drew to the south and east some distance beyond the present shore-line;
the Lower Eocene beds were deposited on a nearly base-leveled surface.
The faunal changes that occurred between the deposition of the
uppermost Cretaceous and the lowermost Eocene strata were very pro-
found; out of 168 species representing 89 genera in the Exogyra costata
zone, which includes the upper part of the Selma chalk (uppermost Cre-
taceous), 20 or more common genera and practically if not all of the
202 : REVIEWS
species became extinct before the Midway group (lowermost Eocene)
was deposited. Stephenson quotes T. W. Vaughan as follows: ‘The
changes that took place in the marine animal life of the Atlantic and
Gulf Coastal Plain during the time represented by the unconformity
separating the Cretaceous and Eocene of this area are more striking than
the changes that have taken place between earliest Midway time and
theipresentidaye asi as
Vo Orr
The Stratigraphy of the Montana Group, with Special Reference to
the Position and Age of the Judith River Formation in North-
Central Montana. By C. F. Bowen. Professional Paper,
U.S. Geol. Survey, No. 90-I, 1915. “‘Shorter Contributions
to General-Geology, 1914" Pp..95=153, plz.
The area treated in this report ‘‘lies east of the Big Snowy and
Judith mountains and extends from Musselshell, on Musselshell River,
to Judith, on Missouri River, Mont.’ The generalized section of the
sedimentary rocks in ascending order is as follows: Cretaceous: Colorado
shale (thickness not measured), Montana group (Eagle sandstone)
(200-300 feet), Claggett formation (7oo feet), Judith River formation
(250-500 feet), Bearpaw shale (1,100 feet); and the Eocene(?)
Lance formation (7oo-800 feet). There is no evidence of an uncon-
formity at any horizon in this section.
It is concluded that ‘‘the evidence of the vertebrate fauna, so far as
in the present state of knowledge it has any weight, and the evidence of
the fresh- and brackish-water invertebrates, so far as it is decisive for
accurate time determination, indicate a closer relationship between the
Belly River [of Canada] and Judith River than between either of these
formations and the Lance. This is in accord with the stratigraphic
evidence, which shows conclusively that both the Judith River and
Belly River formations are separated from the Lance by amarine
formation which is of undoubted Cretaceous age.”
VieOr ene
Statistics of the Mineral Production of Alabama for 1913. By
CHarLtES A. ABELE. Geol. Surv. of Alabama, Bull. 15.
University, 1914. Pp. 67.
A compilation from Mineral Resources of the United States.
aXe JD), 1835
REVIEWS 203
The Coalville Coal Field, Utah. By Carrotyt H. WEGEMANN.
U.S. Geol. Survey, Bull. 581-E, 1915. Pp. 24, pls. 6.
The Coalville coal field lies about 30 miles northeast of Salt Lake
City, in the valley of Weber River. High-grade sub-bituminous coal
has been mined here for more than fifty years.
The report covers four townships. The rocks of the district include
some 8,000 feet of shales and sandstones of Colorado and Montana (?)
age, which are folded into a slightly overturned and pitching anticline,
and are unconformably overlain by 1,000 feet or more of Wasatch con-
glomerate. Several transverse and nearly vertical faults of small dis-
placement cut the gently dipping limb of the fold. Both the folding
and the faulting took place chiefly in pre-Wasatch time, when consider-
able erosion was likewise accomplished; but weaker movement of both
types appears to have followed the deposition of the Wasatch beds.
The one productive coal bed, known as the ‘‘ Wasatch”’ bed, varies
from 5 to 12 feet in thickness. This coal compares favorably in quality
with several Wyoming coals. Coal occurs in thinner seams at two other
horizons, 2,000 feet above and 850 feet below the “Wasatch” bed,
respectively. All three horizons are in the Cretaceous system.
Crewe ie
Preliminary Report on the Clay and Shale Deposits of the Province
of Quebec. By J. Kerr. Canada Dept. of Mines, Memoir
64. Ottawa, 1915. Pp. 280-+iv, pls. XXXIV, figs. 13, map 1.
Describes the clay-bearing horizons and groups producing localities
by the age of the clay produced. Particular emphasis is laid on the
Pleistocene clays. A considerable portion of the memoir is devoted to
the technologic aspects of the clay industries.
Ace Ds Bs
“The Pebble Phosphates of Florida.””’, By E. H. SELLARDs. Florida
Geol. Survey, Seventh Annual Report, 1915, pp. 25-116.
In an earlier report this writer has discussed the origin of hard-rock
phosphates, and this paper extends the study to land-pebble and river-
pebble deposits. The land-pebble phosphates are found in the Bone
Valley formation of late Miocene or early Pliocene age. They form a
portion of a basal conglomerate laid down by a sea advancing over the
Alum Bluff formation, a phosphatic marl of late Oligocene age. In this
204 REVIEWS
marl the phosphate was in the form of pebbles and more finely divided
material. As the result of a long period of erosion covering most of the
Miocene, the phosphatic materials accumulated at the surface and were
reworked by the Bone Valley sea. The river-pebble deposits have
been formed in the beds of streams that have lowered their channels
into either the Alum Bluff or the Bone Valley phosphate horizons.
W. B. W.
Lewis and Gilmer Counties. By DAvip B. REGER. West Virginia
Geol. Survey, 1916. Pp. 660, figs. 12, pls. 30, maps 2.
Several volumes each year are added to the excellent county reports
already published by this state. Lewis and Gilmer counties, located
near the center of the state, have large coal deposits and are rich in oil
and gas. Some of the largest gas wells of the Appalachian field were
drilled in Lewis County.
The Pennsylvanian formations do not reach the development in
these counties that is reported from areas to the south and west. The
Pittsburgh seam of the Monongahela series carries the principal coal
reserve, and the oil and gas sands range in age all the way from the
Chemung to the Dunkard series.
W. B. W.
The Montana Group of Northwestern Montana. By E. STEBINGER.
Professional Paper, U.S. Geol. Survey, No. 90-G, 1914.
“Shorter Contributions to General Geology, 1914.” Pp.
61-66, fig. 1. |
The Montana group of northwestern Montana is composed of four
conformable formations which are, in ascending order: the Virgelle
sandstone (220 feet thick) which is chiefly marine, the Two Medicine
formation (1,950 feet thick) which is mainly a fresh-water deposit, the
marine Bearpaw shale (490 feet thick), and the brackish and marine
Horsethief sandstone (360 feet thick). These formations are similar
to those of the Montana group described in southern Alberta by Dawson,
but differ decidedly from those in the central part of Montana. The
Belly River series of southern Alberta is equivalent to the Virgelle sand-
stone plus the Two Medicine formation, and these in turn are equiva-
lent to the Eagle, Claggett, and Judith River formations (of central
Montana) combined.
VO
REVIEWS 205
Supposed Oil-Bearing Areas of South Australia. By ARTHUR
Wave. Geol. Survey of South Australia, Bull. 4, 1915. Pp.
SAN MIoS ae 24 lSar 3.
This bulletin gives the results of field search for oil-bearing horizons
in four areas along the seacoast. The rocks in these areas are largely
pre-Cambrian schists and quartzites either outcropping or concealed .
by a thin covering of Tertiary sediments. On Kangaroo Island are
clastics assigned to the Cambrian, and glacial deposits doubtfully Permo-
carboniferous in age.
The writer finds nothing in the lithology or structure of the strata
that can be interpreted as favorable to oil accumulation. He believes
that fragments of asphaltum found along the shore have come from beds
down-faulted beneath the sea by the great fractures along the southern
edge of the continent.
W. B. W.
Coal Fields of Kittitas County. By E. J. SAUNDERS. Washington
Geolt Survey, Bull.o,) Pp. 204, fies 52)" pls. 38:
Kittitas County has led all the counties of this state in the production
of coal for many years. ‘The output is chiefly from a 19-foot vein in the
Roslyn formation of middle Eocene age. The coal is of good bituminous
grade and quite free from impurities. The general structural relations
of the beds are quite simple, but there are many striking local exceptions,
and some of these are illustrated by an excellent series of photographs
and diagrams.
In certain portions of the county the Manastash formation of upper
Eocene age carries coal beds that are of no commercial importance at
present.
W. B. W.
Contributions to the Stratigraphy of Southwestern Colorado. By
WHITMAN Cross and Esper S. LARSEN. Professional Paper,
U.S. Geol. Survey, No. 90-E, 1914. ‘Shorter Contributions
to General Geology, 1914.”’ Pp. 39-50, pl..1, figs. 2.
The overlap of the Gunnison formation (= La Plata sandstone (Juras-
sic) below -+ McElmo formation (Jurassic : ?| (above) on pre-Cambrian
rocks “‘extends at least 50 miles farther up the valleys of the Gunnison
and Tomichi than was represented for the Jurassic beds on the Hayden
MAD ieee jhe (c The relations in the Piedra Valley suggest that the La
206 REVIEWS
Plata sandstone overlapped earlier sediments and came into contact
with the pre-Cambrian rocks along a general north-and-south line,
crossing the San Juan Mountains area.” The pre-Dakota section of
Piedra Valley is given.
ieee WO.
A Reconnaissance in the Canyon Range, West-Central Utah. By
G. F. Loucuitn. Professional Paper, U.S. Geol. Survey,
No. 90-F, 1914. ‘Contributions to General Geology, 1914.”
Pp. 51-66, pl. 1, figs. 4-8 (including map).
The Canyon Range in west-central Utah is formed almost wholly
of lower Mississippian and older(?) limestone, and upper Mississippian
and Pennsylvanian(?) quartzite, overlain unconformably by Eocene
conglomerate. Pleistocene Lake Bonneville beds and, locally, later
Quaternary deposits floor the valleys on either side of the range. “ Vol-
canic rocks have been reported from the extreme northern and south-
western parts of the range, beyond the limits of the area visited.”
VE Orde
The History of a Portion of Yampa River, Colorado, and Its Possible
Bearing on That of Green River. By E. T. Hancock. Pro-
fessional Paper, U.S. Geol. Survey, No. 90-K, 1915. ‘Shorter
Contributions to General Geology, 1914.”’ Pp. 183-89, pls. 2.
Yampa River, which is one of the principal tributaries of Green River
and empties into it from the east, is believed to be a superimposed stream
whose present course was established after the deposition and emergence
of the Browns Park formation (Eocene ?). It is thought that “the asser-
tions of the antecedent origin of Green River should be accepted only
after more facts have been obtained bearing on the original extent and
thickness of the late Tertiary formations, as well as on the diastrophic
2 : i
history of the Uinta Range. V.0.T.
“The Gold-Bearing Gravels of ‘Beauce Co., Quebec.” By J. B.
TyRRELL. Bull. Am. Inst. Mining Eng., 1915, pp. I-12.
Describes the placer deposits of the Chaudiere, Gilbert, and Des
Plantes rivers. A point of physiographic interest is the suggestion of
an Appalachian center of glaciation, from which the ice is thought to
have moved northwestward over the area included in the report.
Jee 1D)5" 153,
REVIEWS 207
The Manufacture of Gasoline and Benzene-Toluene from Petroleum
and Other Hydrocarbons. By W. F. Rittman, C. B. Dutton,
and E. W. Dean. U.S. Bureau of Mines, Bull. 114, Wash-
ington, 1916. Pp. 268, figs. 45, pls. 9, tables 83.
Contains the details of the methods employed by Rittman and his
associates, with the results obtained on both laboratory and factory
scale. The bulletin is of especial importance because it incorporates
the results of experimental work that has been given wide publicity by
the press.
The demand for the bulletin has been so great that the edition for
free distribution was exhausted within a month of the date of its release
by the Bureau.
Xe, AD aii 8p
‘“An Arrangement of Minerals according to Their Occurrence,”
By E. T. WuHerry and S. T. Gorpon. Proceedings of the
Academy of Natural Sciences of Philadelphia, August, 1915,
Pp. 426-57.
The classification is the most comprehensive attempt that has come
to the notice of the reviewer, and likewise the most successful. ‘The
divisions made are rather too numerous for use in an elementary class,
but are of great value to advanced students. Doubtless other divisions
could be made, which might be more useful for specific studies, such as
a further division of the hydrothermal deposits for studies of ore deposits,
but in general the classification is an improvement over former attempts.
AD eB.
Corundum, Its Occurrence, Distribution, Exploitation and Uses.
By A. E. Bartow. Canada Dept. of Mines, Memoir 57.
Pp. 377-+vii, pls. xxviii, fig. 1, maps 2.
Corundum-bearing syenites, nephelite syenites, syenite pegmatites,
and anorthosites occur in three belts north of Lake Onatario. These
rocks are chiefly in the Laurentian gneiss, but are also found cutting the
Grenville series. ‘The memoir is devoted to a detailed description of the
more important localities, including analyses and petrographic descrip-
tion of the rocks, and to the economic and technologic features of the
corundum industry, not only of Canada, but of the industry in various
parts of the world.
A. D. B.
208 REVIEWS
Transactions of the American Institute of Mining Engineers, Vol.
XLVI. New York, rotse.Pp::753.
This volume contains papers and discussions of the New York meet-
ing of February, 1914. It contains papers of interest to mining men,
geologists, metallurgists, oil producers, mill operators. The discussion
of the general topic of revision of mining law occupies a considerable
portion of the volume. More attention is given to the discussion of
petroleum than in any previous volume. Of particular interest in this
connection are papers by von Hofer and by Coste, presenting diametri-
cally opposed views on the origin of petroleum. These, with the discus-
sion they brought out, are worthy of the attention of anyone interested
in the geology of petroleum deposits.
jae Da) 1),
Transactions of the American Institute of Mining Engineers. Vol.
XLIX. New York, 1915. -Pp. 853.
Contains papers and discussions of the Salt Lake City meeting of
August, 1914. The greater number of papers bear on mining, milling,
and metallurgy, especially on the leaching of copper ores and the methods
of precipitation of copper from the solutions obtained. Forty-nine
titles are included. Many of the papers are illustrated.
Aca. RB:
SIGNIFICANCE IX. INNER REORGANIZATION OF THE JUVENILE
III. Tot DEcIsIvE TESTIMONY OF CERTAIN VES- EARTH
ss TIGES OF THE SOLAR SYSTEM X. HIGHER ORGANIZATION IN THE GREAT Con-
IV. Furire Errorts : Tact Horizons
Of Form Letters, Price Lists, Bills, Invoices, Drawings,
Menus, Reports, Anything?
Then take advantage of our offer of
TEN DAYS’ TRIAL, WITHOUT DEPOSIT
and become one of the thousands of satisfied customers who all
agree that Daus Improved Tip Top Duplicator is the simplest,
easiest, and quickest method of duplicating on the market. Our
negative rolls now have our new “‘Dausco” Oiled Parchment
Back, giving additional strength and efficiency 100 Copies
from Pen-written and 50 Copies from Typew 1 e1 Original.
If you have tried other duplicators without success, you will
be more than pleased with ours.
Complete Duplicator, Cap Size, Price $7.50
é
ve co
FELIX H. DAUS DUPLICATOR COMP
IMPORTANT NEW NOVELS
By H. G. Wells, J. C. Snaith, Sir Gilbert Parker, Coningsby Dawson, Booth Tarkington, William
J. Locke, F. Hopkinson Smith, and other distinguished American and English writers have been
published recently.
(Prints 834x13 inches) . .. .
ANY, Daus Building, 111 John Street, New York
Descriptions of these new novels and other important recent publications are contained in the
December issue of our
The Monthly Bulletin of the Latest and Best Books
Sent on request and without charge to anyone interested
fd
THE BAKER & TAYLOR CO.
Wholesale Dealers in the Books of All Publishers
354 Fourth Avenue NEW YORK At Twenty-sixth Street
OF THE EARTH
By THOMAS C. CHAMBERLIN
Head of the Department of Geology, The University of Chicago
SEE eee eee
xii+ 272 pages, 8vo, cloth : $1.50 net, postage extra (weight 1 lb. 6 oz.)
FROM THE PREFACE
“Tn ‘telling the story of this search for the mode by which the earth came into being, we
have let the incidents that led the inquiry on from one stage to another fall in with the steps of
the inquiry itself. It is in keeping with the purposes of this series of booklets that the motives
which set researches going should have their place with the quests that arose from them. .. . .
The final story of the birth of the earth will come only after a time, when the vestiges of creation
have been more keenly discerned and more faithfully rendered than is possible now.”
CONTENTS
INTRODUCTION ; V. THE FORBIDDEN FIELD
‘CHAPTER VI. Dynamic ENCOUNTER BY CLOSE APPROACH
I. Tue GAsrous THEORY OF EARTH-GENESIS IN VII. THE EVOLUTION OF THE SOLAR NEBULA INTO
THE LIGHT OF THE KINETIC THEORY OF GASES THE PLANETARY SYSTEM
II. VEsTIGES OF CoSMOGONIC STATES AND THEIR VIII. THE JUVENILE SHAPING OF THE EARTH
One of ‘‘The University of Chicago Science Series’’
THE UNIVERSITY OF CHICAGO PRESS
CHICAGO, ILLINOIS
| FINE INKS 4x» ADHESIVES
For those who KNOW
| Just the Typewriter |
one for Yous. “sam
SOLD ON JUST THE
TERMS THAT SUIT YOU
The Remington J unior
Drawing Inks
Eternal Writing Ink
Engrossing Ink
2 8 9 Taurine Mucilage
ri | io g Ins Photo Mounter Paste
Drawing Board Paste
Liquid Paste
Office Paste
Vegetable Glue, Etc.
Are the Finest and Best Inks and Adhesives
Emancipate yourself from the use of corrosive and
ill-smelling inks and adhesives and adopt the
Higgins Inks and Adhesives. They will bea
revelation to you, they are so sweet, clean, well
put up, and withal so efficient.
At Dealers Generally
The Junior is smaller and lighter than the Stand- _
ard Remington models—weighs only 17 pounds.
It is simpler. You can quickly learn to operate
it. No lessons needed.
It has all the Remington essentials, standard key-
board, standard type, and writes letters of stand- —
ard size—the kind with the hundred-dollar look.
Cut out and post today
REMINGTON TYPEWRITER COMPANY
(Incorporated)
327 Broadway, New York
Send me a Remington Junior Typewriter, price $50, on free ex-
amination. It is understood that I may return the machine, if —
I choose, within ten days. If I decide to purchase it, I agree to
pay for itin 10 monthly payments of $5 each.
| CHAS. M. HIGGINS & CO., Mfrs.
Branches: Chicago, London
| 271 Ninth Street Brooklyn, N.Y.
Optical efficiency, durability, ease of
operation—these are but a few of the
integral features which sixty years of |
experience as manufacturers of optical
products have taught us toembodyinthe
Bausch |omb
BALOPTICON
THE PERFECT STEREOPTICON
The optical equipment of the Balopticon is far in advance of the usual projection lantern.
Model C (illustrated) has the new gas-filled Mazda lamp, which furnishes more evenly
distributed light than the old-style A.C. arc on less current. Takes all standard slides.
Price $42.00. Models for the projection of opaque objects (photos, specimens, maps, etc.)
$35.00 up. Combination models for both forms, with instant interchange $45.00 up.
Illustrated price-list giving full description of different models and their uses, free
on request.
Bausch fF lomb Optical ©.
554 S™ PAUL STREET ROCHESTER, N.Y
NEW YORK WASHINGTON CHICAGO SAN FRANCISCO
Leading American Makers of Microscopes, Projection Apparatus, Photographic Lenses, Engineering Instruments,
Stereo-Prism Binoculars, and other High-Grade Optical Products ‘
NUMBER 3
THE
JOURNAL or GEOLOGY
A SEMI-QUARTERLY
EDITED BY
THOMAS C. CHAMBERLIN AND ROLLIN D. SALISBURY
With the Active Collaboration of a
hi SAMUEL W. WILLISTON, Vertebrate Paleontology ALBERT JOHANNSEN, Petrology
STUART WELLER, Invertebrate Paleontology ROLLIN T. CHAMBERLIN, Dynamic Geology
Hrs ALBERT D. BROKAW, Economic Geology
yy ASSOCIATE EDITORS
ARCHIBALD GEIKIE, Great Britain JOSEPH P.IDDINGS, Washington, D.C.
CHARLES BARROIS, France JOHN C. BRANNER, Leland Stanford Junior University
BRECHT PENCK, Germany RICHARD A. F. PENROSE, Jr., Philadelphia, Pa.
NS REUSCH, Norway _ WILLIAM B. CLARK, Johns Hopkins University
SERARD DEGEER, Sweden WILLIAM H. HOBBS, University of Michigan
T, W. EDGEWORTH DAVID, Australia FRANK D. ADAMS, McGill University
B ILEY WILLIS, Leland Stanford Junior University CHARLES K. LEITH, University of Wisconsin
GROVE K. GILBERT, Washington, D.C. WALLACE W. ATWOOD, Harvard University
; HARLES D. WALCOTT, Smithsonian Institution WILLIAM H. EMMONS, University of Minnesota
NRY S. WILLIAMS, Cornell University ARTHUR L. DAY, Carnegie Institution
we APRIL-MAY 1917
E PROBLEM OF THE ANORTHOSITES TNE fii uk nasi el haat engl es
200
THE MIDDLE PALEOZOIC STRATIGRAPHY OF THE CENTRAL ROCKY MOUNTAIN
pee ele CEC ODIN] pee clad oi clases = rccaminnen ym aha ogee =) “CMAN POMEINSON 244
FEW INTERESTING PHENOMENA ON THE ERUPTION OF USU - Y. Otnouye 258
NTRAFORMATIONAL PEBBLES IN-THE RICHMOND GROUP, AT WINCHESTER,
_ OHIO - - - - - - - - - - - Aucust F. FoERSTE 289
THE UNIVERSITY OF CHICAGO PRESS
CHICAGO, ILLINOIS, U.S.A.
ree : AGENTS:
THE CAMBRIDGE UNIVERSITY PRESS, Lonpon anp EpInsurGH
THE MARUZEN-KABUSHIKI-KAISHA, Toxvo, Osaxa, Kyoto, Fukuoka, SENDAI
_ THE MISSION BOOK COMPANY, Snancuat
pas
THE JOURNAL OF GEOLOGY
EDITED BY
THOMAS C. CHAMBERLIN AND ROLLIN D. SALISBURY
With the Active Collaboration of
SAMUEL W. WILLISTON ALBERT JOHANNSEN
Vertebrate Paleontology Petrology
STUART WELLER ROLLIN T. CHAMBERLIN
Invertebrate Paleontology Banas Geology
ALBERT D, BROKAW
Economic Geology
The Journal of Geology is published by the University of Chicago at the University Press semi-
quarterly, on or about the following dates: February 1, March 15, May 1, June 15, August 1, September 15,
November 1, December 15. {| The subscription price is $4.00 per year; the price of single copies is
65 cents. Orders for service of less than a half-year will be charged at the single-copy rate. J Postage is
prepaid by the publishers on all orders from the United States, Mexico, Cuba, Porto Rico, Panama Canal
Zone, Republic of Panama, Hawaiian Islands, Philippine Islands, Guam, Samoan Islands, Shanghai.
4] Postage is charged extra as follows: For Canada, 30 cents on annual subscriptions (total $4.30), on
single copies, 4 cents (total 69 cents); for all other countries in the Postal Union, 53 cents on annual
subscriptions (total $4.53), on single copies, 11 cents (total 76 cents). [Patrons are requested to make
all remittances payable to the University of Chicago Press in postal or express mony) orders or bank
drafts.
The following agents have been appointed and are authorized to quote the prices indicated:
For the British Empire: The Cambridge University Press, Fetter Lane, London, E.C.4. Yearly
subscriptions, including postage, 18s. 8¢. each; single copies, including postage, 3s. 2d. each.
For Japan and Korea: The Maruzen-Kabushiki-Kaisha, 11 to 16 Nihonbashi Tori Sanchome,
Tokyo, Japan. Yearly subscriptions, including postage, Yen 9.00 each; single copies, including f
postage, Yen 1.50 each.
For China: The Mission Book Company, 18 Peking Road, Shanghai. Yearly subscriptions,
$4.00; single copies, 65 cents, or their equivalents in Chinese money. Postage extra, if mailed direct
outside of Shanghai, on yearly subscriptions, 53 cents, on single copies, 11 cents.
Claims for missing numbers should be made within the month following the regular month of
publication. The publishers expect to supply missing numbers free only when losses have been sustained
in transit, and when the reserve stock will permit. Deliveries are not guaranteed within the war zones pe
Europe, nor in other countries where transportation is disturbed by conditions incident to war.
Business correspondence should be addressed to The University of Chicago Press, Chicago, Ill.
Communications for the editors and manuscripts should be addressed to the Editors of THE
JOURNAL OF GEOLOGY, the University of Chicago, Chicago, Il.
Contributors are requested to write scientific and proper names with particular care and in citations ©
to follow the form shown in the pages of the Journal.
Separates, if desired, must be ordered in advance of publication. Separates of leading articles will be
printed as ordered, of which 50 (without covers) will be furnished gratis, the actual cost of the remainder
(and covers, if desired) to be paid for by the author. The actual cost may vary, and will depend upon __
the amount of work in re-making the pages into forms, presswork, paper, binding, etc.. Separates
containing half-tones may be expected to cost somewhat more, the increase depending upon the number
of cuts and the amount of work required upon them.
Entered as second-class matter, March 20, 1893, at the Post-office at Chicago, Ill., under the Act of March 3, 1879
VOLUME XXV NUMBER 3
THE
ROURNAL OF GEOLOGY
APRIL-MAY 1917
THE PROBLEM OF THE ANORTHOSITES
N. L. BOWEN
Geophysical Laboratory, Carnegie Institution of Washington
STATEMENT OF THE PROBLEM
Seldom can it be truly said that the puzzling feature of any
object is its simplicity, yet of all the problems that the anorthosites
present to us for solution the most difficult is their simple minera-
logical composition. Bunsen long ago taught geologists to think of
rock magmas as solutions, and the so-called solution theory™ of
magmas has now gained general acceptance. We have been
enabled to understand many features of magmas that without the
aid of the theory of solutions would remain incomprehensible. We
understand why the order of separation of mineral from a magma
is not simply the order of their fusibility. We understand also why
a rock magma remains liquid at temperatures far below the tem-
peratures of fusion of the individual minerals that enter into the
magma and, therefore, why magmatic temperatures are. compara-
tively moderate. It is because the individual minerals exist in the
magma in mutual solution and therefore have their specific
properties modified. But when we turn to anorthosites we find
1 Nowadays it is scarcely proper to speak of the solution theory of magmas, for
magmas are solution in virtue of the definition of a solution rather than by theory.
On the other hand, to speak of the theory of solutions as applied to magmas is, of
course, entirely permissible.
209
210 N. L. BOWEN
them made up almost exclusively of the single mineral, plagioclase.
What, then, of the magmatic temperatures of anorthosites? Have
we in them an exception to the rule of moderate magmatic
temperatures ?
Anorthosites apparently exert no exceptional influence upon
surrounding rocks. Foreign inclusions, even very susceptible ones,
are apparently not melted up. Inclusions of quartz-bearing rocks
do not have their quartz changed to tridymite or cristobalite.
Nothing in the field evidence gives us any reason to believe that
anorthosites are in any way exceptional in this respect. Neither
do we find any comfort in field evidence if we entertain the possi-
bility that, in the absence of other minerals in amounts adequate
to produce a great lowering of the melting temperature of plagio-
clase, there was present instead a sufficient amount of the much
more potent “‘mineralizers.”” Typical anorthosite is notably free
from all those minerals which, when present in rocks, constitute the
principal evidence of the presence of volatile components in signifi-
cant amounts in their magmas."
An alternative possibility is that the material of anorthosites
actually was in solution in something else at one time and that it
differentiated from this solution. This alternative is more in har-
mony with general opinion, for few would state that beneath those
places where we find anorthosites, there existed some anorthosite
magma and that it simply was always there. It is generally
believed, rather, that anorthosites are differentiates of gabbroid
magma, the belief being based on field association. But it is also
commonly believed that the differentiation took place in some
manner in the liquid state and produced anorthosite magma.
Now we must realize, and face the fact squarely, that anorthosite
magma, however produced, is, nevertheless, anorthosite magma,
and must exhibit the appropriate characteristics. It could separate
as a liquid, by any process whatsoever, only at temperatures at
which it could exist as a liquid, and we are immediately presented
with precisely the same temperature problem. Anorthosites, as
we have seen, do not give evidence of ever being at a temperature
approaching that requisite to melt plagioclase.
‘This matter is considered in greater detail in connection with the Morin
anorthosite.
THE PROBLEM OF THE ANORTHOSITES 201i
THE PROPOSED SOLUTION
For this reason in part, but also because careful consideration
of all the possibilities and much experimental work to test these
possibilities seem to indicate the inadequacy of any process other
than crystallization, it is believed that the gabbroid magma must
proceed to crystallization, and that anorthosite masses are simply
collected plagioclase crystals. It is believed, then, that anorthosites
were never liquid as such, but that their material when liquid was
part of a solution probably of a gabbroid nature. Only in virtue
of the sorting of solid, crystalline units from this solution does
anorthosite come into being.
Having arrived at this belief, we may examine the anorthosites
to see in how far they agree with its consequences, but before this
can be done it is necessary to discuss in detail the process of sorting
of crystals.
THE PROCESS OF ACCUMULATION OF CRYSTALS
General relations involved—It must be admitted that the
problem of the method of accumulation of plagioclase crystals for
the formation of a mass of anorthosite is not a simple one. If the
plagioclase crystals were much heavier or much lighter than gab-
broic magma, all would be plain sailing. It could then be assumed
that the crystals sank in the magma or floated in it, and were there-
fore accumulated at the bottom or at the top. But laboratory
determinations of the densities of calcic plagioclase crystals and of
molten gabbro place them very close together, with the crystals a
very little lighter, and while the difference would not be the same
and might even be in the opposite direction, there is, nevertheless,
every reason to believe that it would still be small under natural
conditions. Asa matter of fact, we actually find this similarity of
density expressed in the composition of anorthosites. If plagio-
clase were much lighter or much heavier, the mere accumulation of
crystals would be, as we have seen, a simple matter, but the
anorthosite masses formed would not be such as we find them, a
fact which will become obvious from the following considerations.
During the crystallization of a magma involving the precipita-
tion of mix-crystals the crystals first deposited are rich in the higher-
212 N. L. BOWEN
melting component of the mix-crystal series. As the magma cools,
especially if it cools very slowly, these crystals continually change
in composition as a result of interchange of material between liquid
and crystals, the change being always in the direction of enrich-
ment in the lower-melting component. But this change is fully
accomplished only when adequate liquid is available. If the
crystals are heavy and accumulate toward the bottom, the small
amount of liquid there available cannot continue indefinitely to
enrich the crystals in the lower-melting component. They there-
fore remain very rich in the higher-melting component, increasingly
so the greater the preliminary accumulation of crystals. Vogt’s
important statistical study of the anchi-monomineralic rocks shows
quite definitely that in the case of peridotites the ratio of Mg,Si0,
to Fe,SiO, in the olivine increases directly with the proportion of
olivine in the rock.t. The orthorhombic pyroxenes apparently
follow a parallel law.
This is, then, precisely as deduced above for rocks formed by the
accumulation of heavy early crystals. If plagioclase were a very
heavy or a very light mineral, we should find a similar relation to
hold for it, namely, the greater the proportion of plagioclase in a
rock, the greater would be the proportion of anorthite in the plagio-
clase. But when we turn to Vogt’s similar study of anorthosites,
we find in them a quite different tendency. The richer a rock is in
plagioclase, the greater the tendency for the plagioclase to be, not
a very calcic one, but the intermediate one, labradorite.* This
character of the plagioclase is, it is believed, directly connected with
the fact that the plagioclase being precipitated from gabbroic
magma sensibly matches the magma in density. It is perhaps
slightly lighter than the magma, but usually not sufficiently so to
cause it to accumulate locally and to form masses of crystals much
enriched in the higher-melting component as do olivine and
pyroxene. Instead, it remains practically suspended in the liquid,
with probably a very slight tendency to rise at first, and the whole
of the liquid is available for the production of the change of com-
«J. H. L. Vogt, “Uber anchi-monomineralische und anchi-eutektische Eruptiv-
gesteine,” Vid. Selsk. Skr. I, No. 10 (1908), pp. 24-25.
2 Vogt, op. cil., p. 41.
THE PROBLEM OF THE ANORTHOSITES 213
position that ensues as the temperature falls. Thus, though the
earlier crystals of plagioclase are basic bytownite, they are, in
nearly all cases, gradually made over into labradorite by the liquid
in which they remain suspended. In the meantime the liquid has
suffered impoverishment in ferromagnesian constituents and
eventually becomes decisively lighter than the plagioclase crystals.
Then and then only, as a rule, does subsidence of plagioclase
crystals become an important factor, and masses of anorthosite,
anorthosite-gabbro, etc., are formed according to the degree of con-
centration of crystals. It is to be noted that this lighter liquid
from which the labradorite crystals accumulate is now, of course,
no longer gabbroic, but, as a result of removal of femic constituents
and plagioclase, it approaches syenitic composition, and with con-
tinuation of the process actually attains the composition of syenite
or granite. In the ideal case in which the process had free scope
the resultant mass would be stratified, and would consist of syenite-
granite, anorthosite, and pyroxenite in descending order with, in
some cases, peridotites at the base. Of all these the only type that
was ever liquid as such would be the syenite-granite, though liquids
oi every composition intermediate between that of the original
gabbro and the syenite would be concerned in the process and might
occur as chilled borders or in satellitic bodies. The anorthosites
should be intimately related to gabbro, therefore, but as intimately °
related to syenite also, which might occur as interstitial material of
late crystallization in some of the phases. By increase of this
interstitial material gradual transition into syenite might occur.
Intimate field association of anorthosite with gabbro and with
syenite is a fact that no one will question. Some have emphasized
its relation to gabbro and some that to syenite, but the emphasis
is due as much to the personal element and to the kind of exposures
in any particular area as to any fundamental difference between the
anorthosites of one area and of another. One would expéct, to be
sure, that the andesine-labradorite phase of an anorthosite mass
would be the more intimately related to syenite, and the labradorite-
bytownite phase to gabbro.
Quantitative considerations.—We can perhaps form a better idea
of the quantitative relations involved in the process of collection of
214 N. L. BOWEN
crystals if we examine the crystallization of mixtures of the system
—diopside, anorthite, albite—not because proportions will be the
same, but because they will be rather of the same order in the
natural system and will aid us in deciding whether the process is a
reasonable method of producing anorthosites as we find them.
DIOPSIDE
c\
ALBITE Ti Ab,AN, Leia ANORTHITE
Fic. 1.—Diagram of crystallization in plagioclase-diopside melts
A liquid of composition F (Fig. 1), which contains 50 per cent
diopside and 50 per cent labradorite (Ab,An,), begins to crystallize
at 1275°, diopside separating first. As the temperature falls the
composition of the liquid changes from F along AFG (directly away
from diopside), and when the temperature 1235° is reached the mass
consists of 17 per cent diopside crystals and 83 per cent liquid of
composition G. At this temperature bytownite of composition
(approximately Ab,An,) begins to crystallize and the composition
THE PROBLEM OF THE ANORTHOSITES 215
of the liquid changes along the boundary curve DE toward D.
Both diopside and plagioclase continue to separate, and the plagio-
clase crystals, not only those separating at any instant, but also
those which had formerly separated, continually change in com-
position, becoming richer in albite. At 1220° the whole mass is
made up of 37 per cent diopside crystals, 25 per cent labradorite
crystals of composition L (Ab;An,), and 38 per cent liquid of
composition A. As the temperature falls still lower the liquid
gradually decreases in amount and continually changes in compo-
sition until at 1200” it is all used up, the last minute quantity having
the composition M. In the meantime diopside and plagioclase
crystals have been separating, and the plagioclase has been changing
continuously in composition until at 1200°, when the last of the
liquid disappears, the composition of the feldspar is Ab;An;. The
whole mass now consists of 50 per cent diopside and 50 per cent
Ab;An,.
Crystallization takes place according to the foregoing outline if
no sinking of crystals occurs. If diopside crystals sink, no effect on
the composition of the liquid results. We should have then at
1220. a mass in which the diopside crystals were of increasing con-
centration toward the bottom, and in which a certain upper portion
was free from diopside crystals, consisting of 60 per cent of liquid
of composition K and 4o per cent labradorite crystals of composi-
tion L (Ab,An,). Let us imagine that at this stage appreciable
sinking of plagioclase crystals begins, and that it increases in
importance as the liquid changes toward M, and therefore becomes
lighter. It is necessary to imagine also that the plagioclase crystals
sink very slowly, and are outstripped by the heavy diopside crystals
which are forming simultaneously and which increase in size more
rapidly since the liquid is becoming relatively impoverished in
diopside. It seems possible, then, that, locally at least, plagioclase
crystals might accumulate in a mass free from diopside crystals and
containing only a little interstitial liquid whose composition would
lie between K and M. If the mass had 20 per cent interstitial
liquid of composition V and 80 per cent crystals slightly more calcic
than Ab,An,, the final rock formed on solidification of the interstitial
liquid ywould’/consist ‘of '95. per cent’ Ab,An, and’ 5» per: cent
216 N. L. BOWEN
diopside (J). If the mass had only 1o per cent interstitial liquid
of composition W and go per cent crystals slightly more calcic than
Ab,An,, the final rock would consist of 98 per cent Ab,;An, and 2
per cent diopside (Z). This degree of concentration of plagioclase
is ample for the production of nearly all anorthosites, and is more
than sufficient for most of them. To understand the formation of
anorthosites of extreme purity it is necessary to follow what has
been happening to the liquid in the meantime, that is, to that part
of the liquid from which crystals have subsided. Instead of becom-
ing completely crystalline when the composition M is attained, it
continues to change its composition toward S and may go even
beyond 5S, that is, it becomes very rich in albite (haplo-syenitic)*
and the diopside becomes a vanishing quantity. Now it can be
considered that locally the interstitial liquid occurring between the
plagioclase crystals was of this type, which it might be if the rate of
interchange of material did not quite keep up to equilibrium require-
ments, a very likely possibility. On complete solidification of such
a mass anorthosite of extreme purity would result. This would be
more likely to give a rock made up of acid labradorite or even of
andesine-labradorite. It is because plagioclase crystals may
accumulate, under certain circumstances, in a liquid which 1s itself
nearly pure plagioclase (though very different from the crystals in
composition) that we can get plagioclase rocks of such extreme
purity.2 For the case of natural rocks the interstitial liquid is
enriched, not merely in albite, but also in orthoclase and to some
extent in quartz. In those anorthosites that run very low in
bisilicates we therefore commonly find 5 per cent or more ortho-
clase, and occasionally some quartz.
In the foregoing the writer has done his best to picture a process
whereby plagioclase crystals may accumulate in sufficient force to
give a mass of anorthosite. Unquestionably there are some
difficulties, the gravest being that connected with the nearly com-
plete sorting of plagioclase and pyroxene, whose periods of crystal-
tN. L. Bowen, “‘The Crystallization of Haplobasaltic, Haplodioritic and Related
Magmas,”’ Am. Jour. Sci. (4), XL (1915), 161.
2 Another possible method of obtaining extreme monomineralic composition is
suggested later (p. 238).
THE PROBLEM OF THE ANORTHOSITES 217,
lization are in large part contemporaneous. If one could assume
that plagioclase follows pyroxene in the crystallization of gabbro, as
some geologists appear to do, the sorting would be a simple matter,
but chemical considerations will not permit such an assumption.
Yet the difficulties do not seem insurmountable, especially in com-
parison with those connected with other processes. Diffusion is
hopelessly incompetent even if it is assumed that its tendency is in
the proper direction. Liquid immiscibility, whose operation in the
case of silicates has nothing to support it, would certainly not tend
to produce pure liquids in any case, but only to produce liquids of
contrasted composition, all being, nevertheless, mutual solutions of
minerals. Added to these is the temperature objection to which
reference was made on an earlier page, and still others might be
mentioned. On the other hand, it does seem reasonably probable
that a mass of gabbroid magma might cool sufficiently slowly to
permit the necessary amount of sorting of crystals especially if it
was a large mass, or if it was very deeply buried.
CHARACTERISTICS OF ANORTHOSITE CONSEQUENT UPON THE
SUPPOSED METHOD OF FORMATION
It must be admitted, however, that opinion as to whether the
process can take place is not a very decisive matter. More impor-
tant is the deduction of its consequences followed by a survey of
the characteristics of anorthosites in order to determine to what
extent they agree with the requirements. If anorthosites are
generated only by the accumulation of crystals, then the more
nearly a rock mass approaches an exclusively plagioclase composi-
tion, the more nearly it should have approached the completely
solid condition when that composition was attained. In discussing
artificial melts we have seen that if we have a portion with 80
per cent plagioclase crystals and 20 per cent interstitial liquid it
would, on crystallization, have g5 per cent plagioclase and 5 per cent
diopside. In other words, a rock containing only 5 per cent diopside
could have had, after that total composition had been attained by
the process we are considering, not more than 20 per cent liquid.
A rock containing 10 per cent diopside could have had a maximum
of 35 per cent liquid, and one containing only 2 per cent diopside
218 N. L. BOWEN
could have had not more than to per cent liquid. For natural
melts the figures would not be the same, and the probability is that
the amount of liquid would be relatively somewhat larger on account
of the presence of orthoclase in the liquid. Assuming the figures
to be approximately the same, it seems necessary to believe that a
rock containing 95 per cent or more plagioclase, if it is true that it
is formed by the method outlined, should exhibit certain charac-
teristics that set it apart from such a rock as a granite, which, as
we know well enough, often occurs in the completely molten
condition. When the plagioclase rock is formed in situ, it need
exhibit no features differentiating it from other igneous rocks
except perhaps a marked coarseness of grain. Such a body, while
still containing nearly its maximum of about 20 per cent inter-
stitial liquid, might be moved en masse, though probably not far,
from the position of its original formation, but this movement
would be accompanied by the development of protoclastic structure,
especially about the margins. Since all crystalline igneous rocks
pass through a stage at which they are 80 per cent crystalline, all
are subject to the possibility of the development of similar structures
under parallel conditions. The plagioclase rock differs only in
that it cannot be moved without developing this structure, since if
moved when containing more than 20 per cent liquid the mass
moved has not yet attained the requisite degree of concentration of
plagioclase crystals. Protoclastic structure and granulation should
therefore be perfectly general features of all moved anorthosite
masses and very common features of all anorthosites.
When we come down to the movement of such material in small
masses, it seems impossible that it would be capable of being injected
into small openings in cold country rock—in other words, that it
would form no small dikes in such rocks, though it might occur as
dikelike masses in consanguineous igneous types, being injected
into them at a time when they themselves were not completely
crystalline. Such material should, moreover, be incapable of
occurring as effusive flows.
For the purposes of the foregoing discussion a mass containing
20 per cent interstitial liquid has been arbitrarily chosen. It is a
matter of opinion how much liquid a mass must have in order to
THE PROBLEM OF THE ANORTHOSITES 219
be injected as small dikes. If it is considered that about 50 per
cent liquid is necessary, then only anorthosite or, better, anorthosite-
gabbro, with about 85 per cent plagioclase could occur as small
dikes. If somewhat less than 50 per cent liquid is necessary, then
a rock somewhat richer in plagioclase could occur in that manner.
In the case of effusive masses, if it is considered that more than 50
per cent liquid is normally requisite for their formation, only anor-
thosites with less than 85 per cent plagioclase could occur as
effusives.
A study of the literature of anorthosites from various localities
seems to show that, in so far as published descriptions are con-
cerned, anorthosites do have substantially the characters outlined
in the foregoing discussion, which is based on the hypothesis that
they are accumulated masses of plagioclase crystals. Still the idea
is rather novel, and probably no one had such a hypothesis in mind
when examining anorthosites, so that, while many observations
bearing directly on the problem are recorded, one might readily
believe that perhaps many others equally pertinent escaped record.
For this reason the writer spent a few weeks in the Adirondack area
and in the Morin area of anorthosites, becoming acquainted at first
hand with the relations there found. Attention was confined
almost entirely to parts already mapped in detail so that a maxi-
mum could be seen in the limited time. The facts bearing on the
origin of anorthosites in these areas will be stated principally as
recorded by others, and only to a very limited extent supplemented
by this brief personal experience. It is desired to express thanks to
Professors Kemp, Cushing, and Adams and to Mr. Dresser for
interest taken and for furtherance of the work in various ways.
THE ADIRONDACK ANORTHOSITE
General relations —The anorthosite of the Adirondacks occurs
principally as a single great area, for the most part in the heart of
the mountains and making up its highest peaks, though extending
eastward to the lower country in the vicinity of Lake Champlain.
The mass occupies an area approximating 1,200 square miles, the
principal constituent of the rock throughout this area being
plagioclase. Large exposures may be made up almost exclusively
220 N. L. BOWEN
of plagioclase, while other exposures, perhaps equally general and °
widespread, would average nearly 1o per cent bisilicates. This
latter type seems to prevail even in the heart of the area being rep-
resented in most of the exposures of the Keene Valley, while the
bare ledges of the summit of Mt. Marcy average probably more
than 5 per cent bisilicates. Toward its borders, too, the anortho-
site commonly passes into anorthosite-gabbro and gabbro. Never-
theless, the mass as a whole is aptly described as consisting of
“little else than feldspar which is generally a blue labradorite.’”
If this great mass, whose volume is to be measured in thousands of
cubic miles, was ever molten as such, it is remarkable that none of
the many investigators who have studied the area have found a
single dike consisting of nearly pure plagioclase in the surrounding
rocks. The evidence of the intrusive nature of anorthosite in its
more typical development depends on the occurrence of Grenville
inclusions in it. To be sure, the anorthosite is nearly always
immediately surrounded by a younger rock, the syenite, but in
several localities the invading power of certain phases of the anor-
thosite mass is well shown. Anorthosite-gabbro invades the
Grenville and associated older gneisses in places, and occurs as
outlying masses upward of 20 miles distant from the main anortho-
site mass. As soon, then, as the bisilicates mount to 20 or 25 per
cent there is no lack of evidence of the power of the mixture to
penetrate into openings in the surrounding rocks. The great mass
of the anorthosite itself contains much fewer bisilicates, yet in spite
of the overwhelming volume of such material it is entirely unrepre-
sented as dikes and small intrusions. It seems to be a reasonable
conclusion that this material was incapable of being injected into
the older rocks.
Intimate relation of syenite and anorthosite.—The anorthosite core
of the Adirondack igneous mass is surrounded practically every-
where by the syenite-granite series with which are associated
numerous areas of Grenville sediments and perhaps older granite
gneiss. There has been a considerable tendency to consider the °
syenite-granite as an igneous unit and anorthosite as a separate
unit. This tendency has been emphasized perhaps by the fact
1D. H. Newland, V.Y. State Museum Bull. 119, 1908, p. 17.
THE PROBLEM OF THE ANORTHOSITES 225
that in one locality, in the vicinity of Long Lake, Cushing was able
to demonstrate that the syenite is younger than the anorthosite.
Yet even Cushing states: “‘ The syenite and anorthosite seem surely
derivatives from the same parent magma and of no great difference
in age.’’* ‘This aspect of the anorthosite, 1.e., its intimate connec-
tion with the syenite, is emphasized in the area as a whole, where,
in spite of fairly good exposures, only one other locality showing
the intrusive relation of syenite to anorthosite has been found, but
where, on the other hand, types intermediate between the two are
rather commonly found. This feature of Adirondack igneous
geology has not been studied in detail except, apparently, at the
one locality in the Long Lake quadrangle, though it appears to
deserve such study since it marks the great similarity between the
Adirondack anorthosites and others, the Norwegian and Volhynian
occurrences, for example. In the writer’s limited experience it was
found that the change from anorthosite to syenite was heralded by
the appearance of inclusions of potash feldspar in the plagioclase.
The inclusions are small patches, uniformly oriented and consti-
tuting therefore an antiperthite.2_ These inclusions often show a
rather peculiar feature which, so far as the writer is aware, has not
been noted elsewhere. Surrounding some of them and correspond-
ing in general though not in detail with the outline of the inclusion
is an area of plagioclase differing from the crystal as a whole. Its
outline is usually sufficiently sharp to make it possible to determine
that it has a slightly higher refractive index than the rest of the
plagioclase, besides a different position of extinction which makes
it a rather conspicuous feature. An extremely fine twinning, not
shown by the main body of the plagioclase crystal, can usually be
seen with high magnification. A suggested explanation of these
features is that the material of the microcline inclusions was origi-
nally in solid solution in the plagioclase and that on separating
from solid solution it left the plagioclase poorer in potash feldspar,
and therefore of higher refraction than the general body of the
crystal more remote from the inclusions. But the rims about the
inclusions usually have not much greater mass than the inclusions
* Bull. Geol. Soc. Am., XVIII (1907), 485.
2F. E. Suess, Jahrb. K. K. geol. Reichsanst., LIV (1904), 417-30.
229 N. L. BOWEN
themselves, and it would be necessary for the surrounding plagio-
clase to have contained originally nearly one-half potash feldspar,
which it certainly did not. It seems more likely that the potash
feldspar, though occurring as definite inclusions, was, nevertheless,
formed from the portion which remained liquid last and was intro-
duced into the plagioclase by a sort of replacement, the change in
the plagioclase aureole being an effect going hand in hand with this
replacement.
With the microcline inclusions some interstitial microcline
generally makes its appearance, and this may increase in amount
until it becomes an important constituent of the rock. In sucha
specimen the plagioclase is usually andesine rather than labradorite,
though the large blue labradorites typical of the anorthosites often
occur as phenocryst-like individuals. The rock is definitely inter-
mediate between anorthosite and syenite, though the microcline, in
the few slides examined, has not as marked a tendency to be per-
thitic as it has in the typical syenite. One sees these intermediate
types in some of the exposures about the shores of Lake Placid.
There is apparently a transition between some of the rocks mapped
as gneiss (syenite-granite) and those mapped as anorthosite in that
vicinity.’ Similar intermediate types are found in the vicinity of
Elizabethtown, and as a whole they seem to be closely analogous
to the perthitophyres of Volhynia as described by Chrustschoff,’
and to the Norwegian monzonites described by Kolderup.
An inter pretation of the structural relations of syenite and anortho-
site—While the anorthosite and syenite are evidently closely
related and connected by transitional types, they are usually very
distinctive. There is one aspect of their field relations in which
they are strongly contrasted and with which the writer was
impressed in the field. In the great area of syenite-granite that
surrounds the anorthosite core, areas of Grenville are exceedingly
numerous. In many of the mapped quadrangles it has been neces-
«Map accompanying report by Kemp, ‘‘Geology of the Lake Placid Region,”
N.Y. State Museum Bull. 21, 1898. Since the above was written the writer has been
informed by Professor William J. Miller that, while the transitional relation is shown,
the syenite also sends dikes into the anorthosite.
2 Tschermak’s Min. Petr. Mitt., 9 (1888), p. 470.
3 Bergens Museums Aarbog, No. V (1896), p. 86.
THE PROBLEM OF THE ANORTHOSITES 223
sary to use a color to represent a mixture of Grenville and syenite
that defies separate mapping. Now the manner of occurrence of
the Grenville when found in considerable areas is commonly as a
roof lapping over the syenite and showing only comparatively
moderate dips. One sees this in typical form on the shores of
Lake Champlain immediately north of Port Henry, and Miller has
recently described this relation in widely scattered Adirondack
localities, the large-scale example in the Blue Mountain quadrangle
being of special interest.*. Syenite and Grenville in this relation
are almost constant companions.
The anorthosite areas, on the other hand, are very different. It
could be said with little exaggeration that on passing the borders of
the anorthosite core one encounters only anorthosite. It is true
that inclusions of Grenville have been found, enough to prove the
intrusive nature of the anorthosite, but these appear to be small
completely inclosed blocks and do not suggest actual roof remnants.
In spite of their occasional occurrence the contrast between the
syenite and anorthosite areas is very striking. One has but to
glance at the maps of such areas as the Paradox Lake and Long Lake
quadrangles to be convinced of it. Not only is the anorthosite
unbroken by areas of Grenville, especially away from the margins,
but it is likewise practically free from protrusions of the syenite,
although the syenite is, as we have seen, in part at least, a later
rock. If one pictures the syenite and the anorthosite as conven-
tional batholiths, some difficulty is experienced in accounting for
the foregoing facts. It is necessary to imagine an early intrusion of
a huge plug of anorthosite followed by an intrusion of syenite which
took the form of a hollow cylinder circumscribing it and invading
it only peripherally. All of this must take place without throw-
ing the Grenville series into appressed folds, indeed, without very
significant folding of any kind. It is then necessary to imagine
that erosion removed every vestige of a roof from the small interior
anorthosite area, and left great stretches of it throughout the broad
syenite-granite belt that surrounds it.
All of this is perhaps possible, but at the same time seems highly
improbable. On the other hand, if one pictures the Adirondack
1 William J. Miller, Jour. Geol., XXIV (1916), Sot.
224 N. L. BOWEN
complex as essentially a sheetlike mass with syenite overlying
anorthosite, the facts of Adirondack igneous geology seem to
arrange themselves more rationally. On this supposition one would
expect to find areas of the Grenville roof covering the syenite in
places and to find it relatively little disturbed. In the interior and
Fic. 2.—A. Adirondack complex interpreted as batholitic. -++Anorthosite, \ Syenite
B. Adirondack complex interpreted as laccolithic (undisturbed)
C. Same as B after disturbance. Heavy line indicates erosion surface
eastern region of maximum uplift one would expect to find the
deeper-seated anorthosite laid bare and to find it free from areas of
the roof since it was for the most part separated from the roof by a
layer of syenite. (In Fig. 2 the alternative interpretations of the
Adirondack complex are presented.)
On this supposition of the origin of syenite and anorthosite by
gravitative differentiation of a sheetlike mass it is by no means
THE PROBLEM OF THE ANORTHOSITES 225
necessary that syenite and anorthosite should always grade imper-
ceptibly the one into the other. The Adirondack area is one of
considerable disturbance. It no doubt suffered some disturbance
at the time of the intrusion of these igneous masses, and it has
unquestionably been much faulted since their consolidation. Is it
reasonable to suppose that the region necessarily stood stock-still
during the long period required for the consolidation of these
igneous masses? It is, in fact, likely that faulting took place
during this period as well, and if it occurred at a time when the
anorthosite was completely crystallized but the syenite still molten
then it is quite possible that syenite might thus be brought laterally
against, and acquire an intrusive relation to, anorthosite. The fact
that syenite invades anorthosite locally need not therefore be fatal
to the conception of gravitative differentiation of these two types,
nor does it necessarily indicate the order of their arrival from the
depths. A diagrammatic simplicity is not to be expected, but the
broader relations, including the substantial freedom of the whole
interior. of the anorthosite area from protrusions of syenite, seem
to give a distinct preference to their arrangement substantially as
layers with the syenite above as outlined in the foregoing.’
The writer’s leaning toward differentiation practically in place
as the explanation of the variation of many batholiths has been
criticized publicly by Harker, and by others in private corre-
spondence. It is apparently believed that when one rock invades
another the relation necessarily means that the invading rock
arrived from a deep-seated magma basin subsequent to the other.
This may be quite true, as a rule, but there is little evidence in most
cases that adequate consideration has been given to the alternative
view oi differentiation practically in place with only relatively
minor disturbance during the magmatic period. The petrologist
should be reluctant to reject this possibility without fair trial, for
he destroys some of the hope of solving the problems of igneous
t A few small patches of syenite have been found within the anorthosite area. Ii
these are regarded as having been pushed up from below as pipes, it is rather remark-
able that in no instance has their intrusive nature been demonstrated. On the other
hand, if they are remnants of an overlying syenite, they might well lack a definite
intrusive character.
2 Jour. Geol., XXIV (1916), 554.
226 N. L. BOWEN
geology by thrusting the locus of differentiation ever backward into
unseen depths. The Adirondack intrusives would, it is felt, be
interpreted by many in the conventional manner, and for this
reason some pains have been taken to present the alternative
view.
Daly regards the Adirondack anorthosite-syenite complex as
probably a laccolith. According to his views anorthosites are
formed in laccoliths because those masses suffer little contamina-
tion from wall-rock material and anorthosite is a pure differentiate
of gabbro magma. A certain amount of assimilation of wall rock
can occur, however, without eliminating the possibility of the for-
mation of anorthosite, and under such circumstances the syntectic
magma differentiates in such a manner that syenite is formed. In
batholiths, on the contrary, whose emplacement takes place by
stoping, the consequent assimilation has so important an effect on
the gabbroic magma that no anorthosite is formed, according to
Daly. The writer’s interpretation of the Adirondack complex as a
stratified, sheetlike mass with a lower layer of anorthosite and an
upper layer of syenite intimately associated with the Grenville
sediments is therefore in striking agreement with Daly’s conception.
However, a consideration of crystallizing magmas in the light of
experimental study compels the writer to believe that there would
exist in association with the anorthosite a mass of syenite, even if
the invaded rocks were of infinitely refra¢tory and inert materials.
The mass of syenite was probably augmented by assimilation of
‘foreign rocks, but that is a different matter. Apparently this
opinion is in accord with the conclusions of those best acquainted
with the Adirondack rocks in the field who, while demonstrating
that assimilation takes place, consider it rather as an incident than
as a fundamental factor controlling the genesis of rock types. And,
again, the writer’s interpretation of the igneous mass as sheetlike
is offered merely because of the difficulty of picturing the general
relations otherwise. Indeed, it is not considered that the Adiron-
dack batholith or laccolith or whatever it may be called, is excep-
tional in this respect. Most batholiths are regarded by the writer
as just such masses. Consequently it is believed that the shape of
the intrusive is not the determining factor in the formation of
THE PROBLEM OF THE ANORTHOSITES 22r7
anorthosite. It is rather a balance between density, rate of cooling,
and viscosity such that the necessary amount of sorting of crystals
occurs.
The writer must confess an inability to state precisely the reason
why the species presented in an igneous sequence at one locality may
be different from those at another. It is nevertheless believed that
it is unnecessary that the original magmas need have been different,
or even that the manner of differentiation need have varied. The
results seem to be possible if there was a variation in the extent to
which separation of crystals from liquid and also sorting of indi-
vidual minerals were able to take place. Variations in these factors
depend on physical conditions which have, however, their chemical
consequences, for the removal or non-removal of a crystal has each
a perfectly definite effect on the future course of the liquid.
In one sequence, which is well shown in the pre-Cambrian of
Ontario and in certain British intrusives, there is practically only
gabbro and granite with little that could be described as inter-
mediate. Apparently this is especially likely to be true of masses
of moderate size. In somewhat larger masses ultrabasic rock may
make its appearance as one of the members with occasionally some
anorthosite. Usually for the formation of anorthosite a very large
mass is necessary, and possibly also a deep-seated mass. On the
other hand, for the formation of those sequences that emphasize
intermediate types such as diorite, quartz diorite, and granodiorite
the indications are that very large masses are necessary, but that
they should probably occur at moderate depths. There is nothing
here in the way of hard and fast rules, but there do seem to be
fairly definite tendencies. All of these are reasonably to be con-
sidered the result of differences of the physical conditions under
which cooling took place.
THE MORIN ANORTHOSITE
General features—The Morin anorthosite area of Canada is in
many respects very like the Adirondack aréa. It lies near the edge
of the great pre-Cambrian shield where it is overlapped by Paleo-
zoic rocks. It covers a territory of about 1,000 square miles which,
while not as mountainous as the Adirondacks, is nevertheless quite
228 N. L. BOWEN
rugged, many of the important elevations of the Laurentian
Mountains of that region lying within the boundaries of the anortho-
site mass. As with the Adirondack Mountains, the Laurentian
Mountains have suffered glaciation and lakes abound, with the
result that even in the matter of popularity as a summer resort the
two regions are alike, the Laurentian region drawing a plentiful
supply of tourists on account of its proximity to the Canadian
metropolis. Coming to the more fundamental matters of geologic
structure and petrography we find again a remarkable degree of
similarity to which attention is directed in the sequel.
An area of more than 3,000 square miles comprising the Morin
anorthosite was mapped nearly thirty years ago by F. D. Adams,
and a map published on a scale of four miles to one inch.’ The map
is therefore not as detailed and does not form as useful a guide for
one who would see a great deal in limited time as do the quadrangle
maps of the New York State Museum, which are the result of the
labors of a number of workers. On the other hand, the text of the
report is full of minute descriptions of localities, and a brief visit
was paid to some of these in order to become familiar with them at
first hand.
Relation of anorthosite to the surrounding rocks——The Morin
anorthosite occurs, as does that in the Adirondacks, principally as
a single, great intrusive mass. There is, however, a greater number
of small outlying masses that give the area a somewhat greater
interest with reference to the problem of the origin of anorthosite.
The associated rocks are practically identical with those in the
Adirondacks, consisting of igneous gneisses largely of salic compo-
sition and of sediments of the Grenville series. Of all these Adams
concluded that the anorthosite was the youngest, a relation which
he appears to have considered a general one for the Canadian
anorthosites including the great Saguenay mass. Recent study of
the Saguenay area has shown, however, that there are associated
with the anorthosite certain more salic types, possibly consanguine-
ous with it, but of somewhat later age,? the whole being appar-
t “Geology of a Portion of the Laurentian Area North of the Island of Montreal,”
Geol. Surv. Can. Ann. Rept., Vol. VII, Part J, 1896.
2 Personal communication from Mr. Dresser.
THE PROBLEM OF THE ANORTHOSITES 220
ently a counterpart of the anorthosite-syenite association in the
Adirondacks.
While the writer has nothing very definite to offer concerning a
similar association in the Morin area, certain indications were
found tending to show that detailed study might definitely bring
out its existence there. In the vicinity of Piedmont and extending
southeastward beyond Shawbridge the gneiss which here forms the
southern boundary of the anorthosite mass is a rather fine-grained
greenish rock looking very similar to the syenite of the Adirondacks.
Specimens of this taken at various distances from the borders of
the anorthosite show that it varies considerably. In all cases the
rock is composed principally of plagioclase but, on receding from
the border of the anorthosite, orthoclase continually increases in
importance. There is apparently a perfect transition from anortho-
site toward syenite, though in none of the specimens collected had
the change gone to completion, that is, none of the specimens could
be called typical syenite. In one specimen, however, orthoclase
made up about 30 per cent of the rock, and was accompanied by
some quartz, so that probably the change would not have to be
followed much farther to afford typical syenite.* On account of
this transitional relation it is very difficult, at least where seen by
_the writer, to fix a boundary between anorthosite and gneiss.
Specimens that are apparently typical anorthosite and taken well
within the boundary of the mass as mapped by Adams, are found
to be like those types of anorthosite of the Adirondacks which show
the beginning of transition to syenite in that the plagioclase con-
tains orthoclase inclusions. Specimens from the cliffs north of
Piedmont station show the orthoclase in streaks forming an anti-
perthite much richer in orthoclase than any seen in Adirondack
specimens.? Even specimens taken four miles within the border of
the anorthosite, in the village of Ste. Adéle, show abundant ortho-
clase inclusions in the plagioclase.
« Professor Adams informed the writer in conversation that intermediate mon-
zonite types analogous to the Norwegian types of Kolderup occur in the region, so
that, while they are not described at length in his report, he undoubtedly recognized
such types. ;
2 In none of the Canadian specimens was there seen any peculiar zone of plagioclase
surrounding the orthoclase inclusions as described from the Adirondack localities.
230 N. L. BOWEN
In the vicinity of Piedmont occasional dikelets are seen cutting
the anorthosite, which are found to consist principally of micro-
perthite with some quartz and an unusually large amount of
magnetite, a composition that suggests a syenitic source. Taken
all in all the evidence favors the possibility that we have in the
Morin area syenite and anorthosite related in the same manner as
in the Adirondacks, in part transitional into each other, but the
syenite of somewhat later consolidation. Even in the matter of the
occurrence of a certain aberrant type the two regions show a
further similarity. In the vicinity of Elizabethtown, New York,
there is a peculiar dark rock resembling a basic syenite, but con-
taining phenocrysts of the blue labradorite which is described by
Kemp as the Woolen Mill type.*. This rock is duplicated in both
megascopic appearance and microscopic characters in exposures in
the streets of the village of St. Jérome, Quebec. It is apparently
intimately related to both syenite and anorthosite.
Concerning the structural relations of syenite-granite and
anorthosite it is impossible to say anything definite, since syenite
that may be regarded as probably related to the anorthosite has
not been delimited. About twenty miles east of the anorthosite
mass, syenite-granite makes its appearance from beneath the flat-
lying gneiss of the surrounding country. Adams considers that .
the syenite is much more widespread, the gneiss of the surrounding
area forming merely a relatively thin and little disturbed rooi
over it. If the anorthosite is, as in the Adirondacks, a deep-seated
portion of the same igneous complex, then in order to bring the
anorthosite and the roof gneiss into lateral juxtaposition a con-
siderable movement would be necessary, and it is found that, after
passing westward over a twenty-mile stretch of little disturbance,
the gneiss is then, on approaching the anorthosite, thrown into
sharp folds.2- Moreover, we find on passing within the border of
the anorthosite mass that the typical roof gneiss with its occasional
bands of limestone is absolutely lacking, a fact that suggests that
the roof gneiss was nowhere superposed directly upon the anortho-
«“ Geology of the Elizabethtown and Port Henry Quadrangles,” N.Y. Siate
Museum Bull. 138, 1910, p. 35.
2 Adams, op. cit., pp. 11 and 12.
THE PROBLEM OF THE ANORTHOSITES 231
site. Not impossibly, then, there may be in the Morin area a
stratified mass, made up of syenite above and anorthosite below,
with general relations similar to those we have imagined to exist in
the Adirondacks.
Lack of mineralizers in the Morin anorthosite-—On an earlier
page it was pointed out that there is in anorthosite no supply of
minerals other than plagioclase adequate to produce significant
lowering of the melting temperature of the plagioclase. Anortho-
site could exist as magma, therefore, only at very high temperatures
unless there was present a proportion of volatile components suf-
ficient to produce great lowering. Adam/’s work on the Morin
anorthosite appears to give a definite negative answer to this
possibility. The minerals normal to the anorthosite are those
commonly believed to form from relatively anhydrous melts. The
ferromagnesian material appears typically as pyroxene, not as horn-
blende or mica. There is little if any tendency for the pyroxene to
be made over into hornblende or mica even in the very latest stages
of crystallization when the volatile components would reach their
maximum concentration. Even intense shearing of the rock,
which took place partly during this latest stage of crystallization
and partly immediately subsequent thereto, had no tendency to
develop hornblende and mica from the pyroxene, though under
such conditions it is well known to be particularly susceptible to
this change if there are mineralizers present in significant quantity.
All of the evidence points to a substantial lack of mineralizers.
The Morin anorthosite is in these, as in most respects, typical of
the world’s anorthosites. We are therefore impelled toward the
belief that, inasmuch as anorthosites show no definite high-
temperature characters, they are preferably to be considered as
never having been molten as such.
Anorthosites as small intrusions.—In considering the physical
condition of the anorthosite as bearing on this question of its origin
it is perhaps well to recall the circumstances under which Adams’
investigation was undertaken. Prior thereto there had been a
common tendency to believe that all banded rocks were of sedi-
mentary origin, and since the anorthosite is often markedly banded
it had been regarded as a member of the sedimentary series with
222 N. L. BOWEN
which it is associated. Adams entered the field as the champion
of the newer conception that many banded gneisses are of igneous
origin, and that of these the anorthosite was a prominent repre-
sentative. Under such circumstances it cannot be questioned that
any geologist would search diligently for dikes and tongues of
anorthosite running out into the surrounding rocks, and that
having found them he would not fail to record them. Yet one will
search Adams’ report in vain for a single instance of such a dike.
The wording of the one statement which is an apparent exception |
serves only to emphasize the truth of the above. The anorthosite
mass is described as “‘sending az apophysis’’ into the surrounding
gneiss. The apophysis referred to is a great armlike extension five
miles wide. Attention is directed to this lack of dikes in order to
emphasize that here we have an intrusive of a peculiar character,
not to call in question the interpretation of the anorthosite as an
intrusive. Of that there can be no question. Dikes intimately
related to the anorthosite do occur, but they serve to emphasize
the more that there are none consisting almost entirely of plagio-
clase, though there is mile upon mile of such rock within the main
body of anorthosite. It seems reasonable to conclude, therefore,
from the evidence in the Morin area, that a rock consisting almost
entirely of plagioclase is incapable of being injected as dikes. The
reason for this is to be found, it is believed, in the manner of its
origin, a mass of anorthosite being merely a collected mass of
plagioclase crystals.
There are, as has been stated, several small outlying masses of
anorthosite besides the great central mass. These are listed and
described in detail by Adams. Some of them were visited by the
writer, but nothing need be added, indeed nothing can be added,
to Adams’ statements, which are quite explicit with reference to -
the point that it is desired to emphasize. In discussing the anor-
thosite of the outlying masses in general he states: “It is perhaps
on the whole richer in iron magnesia constituents and often contains
minerals such as hornblende and biotite.’ Statements of like im-
port are made in discussing the bands severally. Of the Kildare
1Qp. cit., p. 116. Italics are the writer’s. 2Tbid-,\p. EL7.
THE PROBLEM OF THE ANORTHOSITES 233
bands he says: ‘‘The rock is on the whole richer in bisilicates than
the Morin anorthosite, approaching more nearly a normal gabbro
or norite in composition.”* Practically the same statement is
made of the Cathcart bands;? and again of the Brandon bands
he says: ‘‘Like most of the small anorthosite bands described in
this report, these from the township of Brandon are usually richer
in bisilicates than a true anorthosite should be.’’3
Apparently, then, these outlying bands always vary from typical
anorthosite, usually toward gabbro, but in one or two instances
perhaps the variation is toward syenite-granite, as suggested by a
content of hornblende and biotite. The bands are by no means
inconsiderable bodies, usually having a width of upward of half a
mile or more and a length of several miles. Even masses of this
size are apparently never made up of nearly pure plagioclase rock,
a fact that accords with the belief that a fair proportion of other
minerals is necessary before anorthosite acquires appreciable
invading power in masses of limited size.
CONSIDERATION OF ANORTHOSITES IN GENERAL
The agreement of the two most completely described areas of
anorthosite on the North American continent with the consequences
of the hypothesis of the origin of anorthosite is apparently rather
good. The Norwegian and the Russian areas are equally signifi-
cant, but no attempt will be made to discuss them in detail. Refer-
ence will be made, however, to the schematic presentation of
differentiation given by Kolderup, which is based entirely on field
evidence, and of which a copy is presented below. Attention
is called to the central position of the norite with its anchi-
monomineralic basic differentiates and its more complex acid
derivatives. These are, it is believed, the accumulations of sorted
crystals on the one hand, and the residual liquids on the other.
Uiibid-. pe 122. 2 [bid., p. 124. 3 Ibid., p. 126.
4A dike of anorthosite in the Cripple Creek country, to which the writer’s atten-
tion was called by Professor Graton as an apparent exception, is described as contain-
ing biotite and quartz. It evidently varies toward granite, and its occurrence as a
dike might reasonably be expected.
234 N. L. BOWEN
KOLDERUP’S REPRESENTATION OF DIFFERENTIATION IN THE EKERSUND-
SOGNDAL ANORTHOSITE AREA
Bronzite granite
Adamellite
Banatite
Monzonite
Quartz-norite
Gabbro-norite—Norite—Anorthosite-norite—Anorthosite
Bronzitite—Norite-bronzitite
Ilmenite-norite
ee
Kolderup’s anorthosites become more basic and schistose toward
their borders and their contact relations are obscure. One cannot
be sure from his text whether there are small dikes of anorthosite
in the surrounding rocks or not, but apparently there are not.
Apophyses consisting of g5 per cent basic labradorite and 5 per
cent augite cut the Sooke gabbro of Vancouver Island described
by Clapp.2_ The anorthosite veins and the gabbro are consanguine-
ous, however, and the former with, say, 20 per cent liquid might
have been squeezed into the not completely crystallized gabbro
mass, the process involved being then rather different from that
occurring in the injection of anorthosite into cold country rock.
Dikes of anorthosite are described as cutting the older rocks in
the Rainy Lake region, but one cannot be sure from the description
whether the dikes are strictly anorthosite or rather the related
anorthosite-gabbro.s And so it is with many descriptions. It is
profitless, therefore, to pursue the discussion of various anorthosite
occurrences further since they were not examined with the questions
« “Tie Labradorfelse des westlichen Norwegens,”’ Bergens Museums Aarbog, No.V
(1896), p. 14.
2C. H. Clapp, “Southern Vancouver Island,” Geol. Survey Canada Mem. No. 13,
1912, p. 116.
3 Since the above was written Professor Coleman has informed me that in so far
as he can recall there are no dikes of typical anorthosite.
THE PROBLEM OF THE ANORTHOSITES 235
raised in mind. An individual cannot do more than state the
problem and leave its suggested solution to confirmation or refuta-
tion at the hands of those acquainted with various anorthosites in
the field.
Of anorthosite in general it can be said, however, that no
effusive equivalent has hitherto been found anywhere. This must
be regarded as a very surprising fact if there were ever masses of
molten plagioclase adequate to furnish such great exposures of
anorthosite as occur in various parts of the earth. On the other
hand, if these anorthosite masses were merely collections of plagio-
clase crystals effusive anorthosites are scarcely to be regarded as
possible.
MONOMINERALIC ROCKS IN GENERAL
Enough has been said incidentally in the foregoing to make it
clear that the problem of any monomineralic rock is, in its essen-
tials, the same as the problem of the anorthosites. There are no
more promising methods of obtaining pure molten olivine or
pyroxene than there are of obtaining molten plagioclase. On the
other hand, the collection of crystals to give substantially solid
masses of nearly pure olivine or pyroxene does not seem out of the
question.
A survey of the domain of igneous geology lends considerable
support to the possibility that peridotites and pyroxenites are so
generated. In making the test of their occurrence as small dikes
we find that this is perhaps the most characteristic manner of
occurrence of peridotite, but on closer examination it appears that
this fact may be due rather to the elastic nature of the term perido-
tite, which may be applied to a rock containing considerable plagio-
clase, pyroxene, hornblende, or mica, or all of these, as well as its
olivine. Typical dunite, or nearly pure olivine rock, however,
probably does not occur as dikes, if we except, again, its occurrence
in such form in closely related and essentially contemporaneous
igneous rocks. The same statement may be made of rock types
excessively rich in pyroxene. As to the question of their occurrence
as effusive types it is found that peridotite has an effusive equiva-
lent in picrite, but picrite is far from being a pure olivine rock.
Dunite itself has apparently no effusive equivalent. With the
236 N. L. BOWEN
pyroxenites the case is apparently the same. Limburgite and
augitite can scarcely be regarded as monomineralic rocks in the
stricter sense of the term, and that is the only sense in which it can
be used in testing the hypothesis. The presence of both pyroxene
and olivine, of a glassy base and usually of some feldspathoid makes
it clear that these effusive pyroxenites do not constitute an exception
to the rule that the monomineralic rocks do not have effusive
equivalents. Apparently, the facts are in accord, therefore, with
the hypothesis that monomineralic rocks are accumulated masses
of crystals. Mention may be made again here of Vogt’s discovery
that the richer a peridotite is in olivine, the richer the olivine is in
magnesia, a fact which is readily explained on the assumption that
peridotites are made up of accumulated early crystals.
All of the monomineralic rocks often do occur, however, in a
manner which has led a very great number of investigators to
speak of the magmas of these rocks as freely as of the magmas of
any others. This is probably due partly to the fact that the possi-
bility of their origin after the manner here advanced did not occur
to the investigators, but whether this was always the case is a
question that, again, an individual cannot answer. One of the
most remarkable occurrences of anchi-monomineralic rocks, espe-
cially pertinent in the present connection, is that described by
Harker from the islands west of Scotland. As a result of his
minute descriptions an especially favorable opportunity is offered
of discussing these rocks in the light of the present conception of
the origin of monomineralic rocks. The rocks are intricately
banded in such a manner as to lead Harker to suggest the intrusion
of a non-uniform magma, implying apparently a non-uniform
liquid? A difficulty in the way of accepting this interpretation is
that connected with obtaining a non-uniform liquid, especially with
1 While it has been necessary in applying the foregoing tests to set aside anchi-
monomineralic rocks containing a considerable amount of other minerals, it should
not be assumed that there is any essential difference in the method of formation.
These are merely examples in which accumulation of crystals of one kind has not taken
place to quite the same degree and which consequently could have had a consider-
able amount of interstitial liquid.
2“ Geology of the Small Isles of Inverness-shire,” Mem. Geol. Survey Scotland,
1908, Pp. 74.
THE PROBLEM OF THE ANORTHOSITES 237
such extremes of composition. There is no promising method of
doing so. Another difficulty presents itself in the very rapid
changes from one type to another. Even granting some method of
obtaining a heterogeneous liquid, one encounters the problem of
maintaining these sharp contrasts in adjacent liquids, for diffusion,
though unquestionably a slow process, would nevertheless accom-
plish much through moderate distances in the time required for
the cooling of such masses. On the other hand, it seems reasonably
possible both to obtain and to maintain almost any degree of
heterogeneity as a result of the accumulation of crystals. On this
assumption it is necessary to imagine the source of the olivine-rich
types in a portion of the magma reservoir where olivine crystals
had accumulated and of the feldspar-rich types where feldspar
crystals had accumulated. These partly crystalline masses were
thrust into the position where found. The greater the approach to
monomineralic composition, the less liquid there could have been.
In accordance with this conception it is found that in the allivalite
the feldspar crystals are arranged with their elongation in the
direction of flow of the sheets, and that this becomes more marked
the richer the rock is in feldspar. In the case of bands consisting
almost entirely of one mineral, which should have had very little
liquid to lubricate their flow, it is found that characters consequent
upon this are developed. Thus the nearly pure feldspar rock is
described by Harker as strongly fissile and the pure olivine rock as
foliated.‘ Possibly connected with the nearly solid condition of
these rocks as injected is the fact that their intrusion apparently
involved overthrusting, at least it is intimately connected with a
line of overthrusting along which earlier, later, and possibly con-
temporaneous movements took place.
In correspondence with the unusual conditions of formation
and intrusion of these ultra-basic rocks we find them to be scarcely
duplicated elsewhere. The Russian ultra-basic rocks described by
Duparc and Pearce seem to be their nearest relatives. They show a
not dissimilar banding of closely related types and possibly may be
. explained in a like manner. The peridotite dikes are described as
1 Op. cit., pp. 72 and 87.
238 N. L. BOWEN
“‘narfois légérement schisteux.”* It may be noted at this point,
also, that an augitite associated with the perfectly massive alkaline
types of the Ice River district, British Columbia, is described as
having a “‘suggestion of a schistose texture.’ Observations such
as these, though seemingly unimportant, may nevertheless have
considerable importance in connection with the movement of a
mass with very little interstitial iquid. It may well be, also, that
in the movement of a mass with a small amount of interstitial
liquid lies the secret of the formation of some monomineralic
masses of extreme purity. Such movement when it caused a
crushing of crystals at their points of contact would necessarily
imply a flowing away of some liquid. Continuance of this action
might, under certain conditions, result in a squeezing out of the
interstitial liquid as from a sponge.
Rocks made up almost exclusively of albite or oligoclase are
known, but there is usually evidence, if only of a collateral nature,
that solutions have played a prominent part in their formation.
Though often occurring as dikes there is never any reason for
believing that these materials have ever been molten as such.
And so it is with many masses of magnetite, indeed it is not impos-
sible that practically any mineral might occur as dikes having a
similar character and origin. Such an occurrence need not, how-
ever, affect one’s belief that, as a rule, monomineralic rocks are
crystal accumulations analogous to the great anorthosite masses
and having the characteristics corresponding thereto.
It will be noted that nowhere in the foregoing discussion has an
appeal been made to the remelting of the masses of crystals once
accumulated. While the writer would not go the length of stating
that such action never takes place, he would nevertheless con-
sider that it must be of very exceptional occurrence. It has been
shown that the monomineralic rocks are best explained without
the aid of the doctrine of remelting, and many of the broader
generalizations of igneous geology are opposed to it. For example,
the parallelism between sequence of intrusion and sequence of con-
™“TOural du nord I,”’ Mem. soc. phys. et @hist. nat. de Genéve, XXXIV (1902),
Fasc. 2, p. 101.
2 Warren, Allan, and Conner, Am. Jour. Sci. (4), XLIII (1917), 75.
THE PROBLEM OF THE ANORTHOSITES 239
solidation is altogether too close to permit one to consider remelting
an important factor. Remelting would almost certainly destroy
all law and order in this matter. Harker has recently expressed a
belief to the contrary, pointing out that the remelting of a solidified
mass with basic material at the bottom and acid at the top might
take place from the bottom upward." Possibly it might, and in
an undisturbed crust it would realize the common sequence of
intrusion, but in an earth’s crust subject to faulting, folding, and
overthrusting it may be doubted whether any regularity would be
observed. Disturbance of the stratified mass would often put
some of the basic material on a level with or even on a higher
horizon than some of the acid material. The remelting of such a
disturbed mass would not give rise to any significant regularity in
the intrusive sequence.
‘A CONSIDERATION OF THE CRITERIA FOR THE RECOGNITION OF
ONCE MOLTEN ROCKS
If we pass in review the development of ideas concerning igneous
or once molten rocks we find them first clearly recognized in surface
lavas. It was natural that it should be so, for here we have rocks
that, judging from their relations to their surroundings, have evi-
dently flowed as a liquid, and that are being duplicated in flows
from active volcanoes at the present day. Then we find a few
coming to believe that other rocks, usually quite distinct in appear-
ance and occurring as deep-seated masses only bared by erosion,
really are made up of the same material, the difference in appear-
ance being principally due to the difference of conditions under
which solidification took place. After much controversy this belief
gains general acceptance, especially as a result of the accumulation
of facts proving the essential identity of these deep-seated masses
with volcanic flows. Originally, then, it was this correspondence
of plutonic rocks with volcanic rocks that gave geologists the right
to consider them once molten or igneous rocks. Simultaneously
with the development of this view numerous facts corroborative of
it accumulated, important among these being the manner in which
the plutonic masses sent tongues into the surrounding rocks, and
t Journal of Geology, XXIV _ (1916), 556.
240 N. L. BOWEN
the light which the microscope threw on the process of crystalliza-
tion of their mineral constituents, which evidently took place pre-
cisely as it should if they were once molten. Eventually, these
corroborative facts came to be the criteria for the recognition of an
igneous or once molten rock and, at present, in actual practice it is
almost exclusively on the basis of the microscopic structure that a
rock is placed as igneous or not. Thus judged, the monomineralic
rocks are unquestionably to be considered as once molten, but if we
revert to the original criteria we find that in some respects they fail
to qualify. In the matter of sending tongues into surrounding
rocks we find them scarcely typical, and as far as occurrences as
lavas are concerned we find them wholly wanting. This apparent
discrepancy is due to the fact that we have not made our distinc-
tions fine enough. These rocks were formerly molten, but they
were never molten as such. When molten they were part of a
complex solution. Monomineralic rocks therefore afford the
strongest justification for believing that crystallization controls
differentiation. If differentiation took place in magmas wholly
liquid, it would seem that all plutonic rocks should have their
effusive equivalents. An examination of any table of classification
of igneous rocks on a mineralogic basis shows, however, a decisive
tendency for plutonic rocks to vary more widely than do effusives,
especially among basic rocks, and especially in this matter of run-
ning to marked richness in one mineral. This fact would have little
significance if it were a fairly common feature of plutonic rocks to
lack an effusive equivalent, but it becomes of the greatest signifi-
cance in connection with the manner of origin here advocated for
the monomineralic rocks when it is realized that in this respect
the monomineralic rocks stand alone.
VOLUME AND AGE RELATIONS OF MONOMINERALIC ROCKS
Of the monomineralic rocks anorthosite is the only one that
occurs in any great amount. The actual volume of pyroxenite and
peridotite exposed at the surface of the earth appears to be insig-
nificant.. On account of the exceptional period required for the
t Daly’s figures would indicate the order of magnitude (Igneous Rocks and Their
Origin, p. 44).
_ THE PROBLEM OF THE ANORTHOSITES 241
sorting of plagioclase crystals anorthosite can form only from very
large masses of magma that cool with great slowness, or if from
masses of more moderate size these must be deep-seated. The anor-
thosite of the large masses normally belongs below the granitic zone
so that, whether formed in very large bodies or in bodies of more
moderate size, it is an especially deep-seated rock. Peridotites and
pyroxenites by reason of the relative ease of sorting of these heavy
minerals can form from moderate masses and at moderate depths,
and are therefore of widespread occurrence and of general distri-
bution in the geologic column though never exposed in large masses.
Anorthosites, on the other hand, being essentially deep-seated are
exposed only in terranes that have suffered, locally at least, excep-
tionally deep erosion, the pre-Cambrian and perhaps early Paleo-
zoic. According to the writer’s opinion there are probably large
masses of peridotite and pyroxenite, but these have not been
exposed at all for the same reason that anorthosite is exposed only
in the older terranes. These peridotites and pyroxenites are, as it
were, the complements of the granites, which in virtue of their low
density are so abundantly exposed. Many will, no doubt, consider
the opinion that there are large unexpected masses of pyroxenite
and peridotite a pure assumption, and it is quite true that some
assumption must be involved in the formation of opinion concern-
ing inaccessible portions of the earth. Nevertheless, an assumption
based on analogy with many completely accessible bodies showing
density stratification should surely be given a preference over an
assumption, tacit or otherwise, that the kind of rocks exposed in
any body extend downward indefinitely, which is based merely on
lack of evidence, one way or the other, for that particular body.
However this may be, it is certain that anyone who believes that
anorthosite is a differentiate of gabbroid magma, as most petrol-
ogists do, must believe that there is an equivalent amount of
pyroxenite somewhere, and if not exposed then presumably in
inaccessible regions. At this point the hypothesis of crystal
accumulation steps in with a rational explanation of the not infre-
quent lack of pyroxenite in anorthosite terranes, very difficult to
account for on the doctrine of liquid differentiation. Being an
accumulated mass of crystals, pyroxenites usually remain sub-
242 N.. L. BOWEN
stantially where formed. If liquid they could not fail to be repre-
sented very prominently in all anorthosite terranes, for the liquid
would be freely intruded into overlying rocks at every disturbance
experienced by them.
SUMMARY
Anorthosites are made up almost exclusively of the single
mineral plagioclase, and in virtue of this fact they present a very
special problem in petrogenesis. The conception of the mutal
solution of minerals in the magma and the lowering of melting
temperature consequent thereon is no longer applicable. Yet
anorthosites give no evidence of being abnormal in the matter of
the temperature to which they have been raised, in other words,
they give no evidence of having been raised to the temperature
requisite to melt plagioclase. A possible alternative is that they
may never have been molten as such, and are formed simply by
the collection of crystals from a complex melt, probably gabbroic
magma. ‘This possibility is in harmony with the expectations that
grow out of experimental studies and for this reason a consideration
of the likelihood that anorthosites have originated in the stated
manner becomes imperative.
A consideration of the method whereby accumulation of plagio-
clase crystals might take place leads to the conclusion that the most
promising is the separation by gravity of the femic constituents
from gabbroid magma, while the plagioclase crystals, which are
basic bytownite, remain practically suspended. ‘Then, at a later
stage, when the liquid has become distinctly lighter, having attained
diorite-syenite composition, the plagioclase crystals, which are now
labradorite, accumulate by sinking and give masses of anorthosite,
at the same time leaving the liquid out of which they settle of a
syenitic or granitic composition.
Some of the consequences of this manner of origin of anortho-
site are as follows. Typical anorthosite, very poor in bisilicates,
should not occur as small dikes, for a mass of accumulated crystals
should have little invading power. A proportion of about 15 or 20
per cent bisilicates or other foreign material such as orthoclase and
quartz should be necessary for the formation of small dikes.
Typical anorthosite should for like reasons not occur as an effusive
THE PROBLEM OF THE ANORTHOSITES 243
rock, a rather large proportion of minerals other than plagioclase
being necessary before such an occurrence would become possible.
Anorthosite should be intimately associated with gabbro, but per-
haps as intimately with syenite or granite. Anorthosites should
commonly be labradorite rocks rather than bytownite or anorthite
rocks.
A consideration of anorthosites with special reference to the
Adirondack and Morin areas gives some reason for believing that
anorthosites do show the requisite characters. For the Adirondack
area especially, evidence is adduced favoring the possibility that
there anorthosite and syenite may still occupy the relative positions
in which they were generated by the process outlined, the Adiron-
dack complex being interpreted as a sheetlike mass with syenite
above and anorthosite below.
Other monomineralic rocks present essentially the same problem
and are restricted in their occurrence in substantially the same
manner if we consider especially those that approach most closely
to the strictly one-mineral character. All of the monomineralic
rocks do occur, however, as dikes and dikelike masses in essentially
contemporaneous, congeneric, igneous rocks, a fact that may be
interpreted as due to the intrusion of a heterogeneous, partly
crystalline mass.
On the whole the inquiry gives considerable support to the
belief that the monomineralic rocks, of which the anorthosites are
perhaps the most important representatives, are generated by the
process of collection of crystals under the action of gravity.
THE MIDDLE PALEOZOIC STRATIGRAPHY OF THE
CENTRAL ROCKY MOUNTAIN REGION
C. W. TOMLINSON
University of Chicago
PART II
STRATIGRAPHY—Continued
UPPER CAMBRIAN AND EARLY ORDOVICIAN
Members o and 1.—Throughout western Wyoming, and at least
as far north as Livingston, Montana, the thin-bedded upper part
of the Gallatin formation rests upon a very massive cliff-making
dolomite (Member 1), ranging up to 400 feet or more in thickness.
Near Three Forks this member constitutes the “mottled limestone’’
(middle division of the Gallatin formation) of Peale," which was
correlated by Weed? with the Pilgrim limestone of the Little Belt
Mountains. It corresponds roughly in position and in general
character to the Hasmark formation, which was differentiated by
- Emmons and Calkins? in the Philipsburg quadrangle, Montana, and
tentatively correlated by them with the Pilgrim.
Underlying this massive member in Wyoming and Montana is
a zone of weaker strata (Member o) in which green shales predomi-
nate, interstratified with thin beds of dolomite and flat-pebble
limestone conglomerate. This belt includes the ‘‘Obdolella shales”’
of Peale? in the Three Forks quadrangle, and at least the upper
part of the Gros Ventre shale of Blackwelder® in western Wyoming;
and is probably represented in the Park shale of Weed® in Montana.
t A.C. Peale, ‘Description of the Three Forks (Montana) Sheet,” Geol. Atlas
U.S., Folio 24 (1896).
2W. H. Weed, “Geology of the Little Belt Mountains, Montana,’ U.S. Geol.
Survey, 20th Ann. Rept., Part 3 (1900), Pl. 40, opp. p. 284; p. 286.
3 W. H. Emmons and F. C. Calkins, “‘Geology and Ore Deposits of the Philips-
burg Quadrangle, Montana,” U.S. Geol. Survey, Prof. Paper 78 (1913), PP. 57-59, 93.
4 Op. cit. 5 Eliot Blackwelder, unpublished manuscript. 6 Op. cit., p. 286.
244
PALEOZOIC STRATIGRAPHY OF ROCKY MOUNTAINS 245
The upper part of the Cambrian sequence in Utah is not unlike
that in Wyoming. There is a very prominent massive member
(No. 1), which is overlain by thin-bedded dolomites (Members
®
ADO
o
4
6
a
4
4
4
a
7
ja
Oo
=
Oo:
O
» House Rang e)
0515 (+?)
2-4) containing a great deal of flat-pebble limestone conglomerate,
and is underlain by a series (Member o) of thin-bedded dolomites
with considerable clastic sediment and a minor constituent of
Fic. 6.—Map showing the extent and thickness of the Upper Cambrian and Early Ordovician series
(Members 1 to 4, inclusive).
246 C. W. TOMLINSON
flat-pebble conglomerate. Walcott™ places the lower boundary of
the Upper Cambrian approximately in the middle of this lower
series (Member o), at the base of a sandstone member. No fossils
have been found in the 1,000 feet of beds (Nounan formation) next
beneath that horizon, but the Bloomington formation, next below,
carries a Middle Cambrian fauna.2 As the shales of Member o in
Montana and Wyoming have been assigned to the Middle Cam-
brian, they are possibly to be correlated with the green shales of
the Bloomington formation.
Members 3 and 4 in northern Utah——The main flat-pebble con-
glomerate series (Member 3) on Blacksmith Fork is 800 feet thick,
and has yielded fossils at several horizons. These collections have
established the Ordovician age of all but the lower 190 feet of the
member, which is left by Walcott? in the Upper Cambrian. The
- Garden City formation in the Randolph quadrangle is described
by Richardson‘ as containing flat-pebble conglomerate throughout
its estimated 1,000 feet of thickness, but it may include a repre-
sentative of Member 4 as well as of Member 3. Richardson’s
collections’ are like those of the writer from Member 4 on Black-
smith Fork, and clearly indicate that this formation is of Beekman-
town age, possibly ranging up into slightly later time.
Richardson® states that he has evidence of an unconformity
between Cambrian and Ordovician, but this evidence is not yet
published. Elsewhere in Utah no evidence has been cited of an
interruption in sedimentation at this horizon, and it is certain, as
indicated by the exact similarity between the latest Cambrian and
the earliest Ordovician rocks, that if similar conditions were not
continuous from the former period into the latter in this region
they returned with very little modification after the interruption.
The upper record succeeds the lower without the intervention of
any clastic sediments.
1C. D. Walcott, “Cambrian Cordilleran Sections,’ Smithsonian Misc. Coll., LYII
(1910), 193.
2 Walcott, op. cit., pp. 194-95. 3 Op. cit., p. I9l.
4G. B. Richardson, ‘The Paleozoic Section in Northern Utah,” Amer. Jour. Sct.,
4th Ser., XXXVI (1913), 406-15.
5 Op. cit., pp. 408-9. 6 Op. cit., p. 408.
PALEOZOIC STRATIGRAPHY OF ROCKY MOUNTAINS 247
Correlation with the Pogonip group of Nevada.—The Pogonip
group of the Eureka district of Nevada, called Lower Silurian by
Hague’ and Walcott,? probably includes equivalents of the Garden
City formation and part or all of the St. Charles formation (Upper
Cambrian, Blacksmith Fork). The upper part of the Pogonip
group carries a fauna, considered by Walcott* to be of Chazyan
age, which bears a notable resemblance to the faunas of the Garden
City formation.
Member 3 in Wyoming and Montana.—In Wyoming the thin-
bedded Upper Cambrian—Lower Ordovician(?) sequence, including
flat-pebble conglomerate, is nowhere represented by more than
500 feet of strata, but its characters are typical of Member 3 wher-
ever it is exposed. In southwestern Montana it forms the highest
member of the Gallatin formation.
The faunas so far collected from this sequence in both Wyoming
and Montana are very meager, a fact which has made its correlation
a subject of dispute in various localities. In the Bighorn Range,
Member 3 constitutes the uppermost part of the Deadwood forma-
tion, which is called by Darton* Middle Cambrian; but the only
species Darton names as coming from the upper 600 feet of the
formation (about 1,000 feet thick in all) is Dicellomus politus, which
is associated in one locality’ with fragments of a trilobite resembling
Ptychoparia owent. Some collections from the upper part of the
Gallatin formation in western Wyoming, comprising three species
of Eoorthis and some fragmentary trilobite remains, have been
referred to the Upper Cambrian by paleontologists of the United
States Geological Survey.6 Member 3 probably is represented
t Arnold Hague, “Geology of the Eureka District, Nevada,’ U.S. Geol. Survey,
Monographs, XX (1892).
2C, D. Walcott, “Paleontology of the Eureka District,” U.S. Geol. Survey,
Monographs, VIII (1884).
3 Op. cit., pp. 3-4.
4N. H. Darton, “Geology of the Bighorn Mountains,” U.S. Geol. Survey, Prof.
Paper 51 (1906), p. 26.
s N. H. Darton, “Fish Remains in Ordovician Rocks in the Bighorn Mountains,
Wyoming, with a Résumé of the Ordovician Geology of the Northwest,” Bull. Geol.
Soc. Amer., XVII (1905), 551.
6 Eliot Blackwelder, personal note. Cf. C.D. Walcott, “Cambrian Brachiopoda,”
U.S. Geol. Survey, Monographs, XXXI (1912), 233, Lot 302¢.
248 C. W. TOMLINSON
in the Red Lion formation (250 feet thick) of the Philipsburg
quadrangle, Montana, from which Kindle" has made some collec-
tions identified by Walcott? as Upper Cambrian.
The beds in question probably are equivalent to some part or
parts of Member 3 of the Utah sequence, and it is therefore possible
that they may include strata of Lower Ordovician age.
The Maxfield formation of the central Wasatch.—Hintze’ has
described a sequence of 481 feet of limestones and shales “‘discon-
formably overlying the Alta shale’”’* on the South Fork of Big
Cottonwood Canyon, southeast of Salt Lake City, which he has
named the Maxfield formation, and has tentatively assigned to
the Ordovician. The Alta shale (1 50-200 feet thick), which tests
on the basal Cambrian quartzite, carries a Lower Cambrian fauna
near its base and a Middle Cambrian fauna at a higher horizon.°
Disconformably above the Maxfield lies the Devonian Benson
limestone.
No fossils have been found in the Maxfield formation. Its
reference to the Ordovician was suggested by the “‘ wormy ”’ appear-
ance of the chief limestone members of the formation, which Hintze
likened to the Lowville (‘‘Birdseye’’) limestone of New York, and
by the occurrence at the top of the formation of ro feet of shale
alternating with typical flat-pebble (‘‘edgewise’’) limestone con-
glomerate, which Hintze® noted had been described from Lower
Ordovician strata elsewhere. Unfortunately for this correlation,
conglomerate of that type is abundantly developed in northern
Utah, not only in the Garden City (Beekmantown) formation, but
in the St. Charles (Upper Cambrian) and Bloomington (Middle
Cambrian) formations. The ‘‘wormy’’ appearance is to be seen
in various members of each and all of the six Middle and Upper
1E. M. Kindle, ‘Fauna and Stratigraphy of the Jefferson Limestone in the
Northern Rocky Mountain Region,” Bull. Amer. Pal., No. 20, 1908, pp. 10-11; also
Emmons and Calkins, op. cit., p. 63.
2C, D. Walcott, op. cit. ult., p. 233, Lots 302q, 302r.
3F. F. Hintze, Jr., “A Contribution to the Geology of the Wasatch Mountains,
Utah,” Annals New York Acad. Sci., XXIII (1913), 85-143.
‘Tbid., p. 105.
5C. D. Walcott, U.S. Geol. Survey, Bull. No. 81, 1891, p. 319.
© Op. cit., p. 106.
PALEOZOIC STRATIGRAPHY OF ROCKY MOUNTAINS 2409
Cambrian formations at Blacksmith Fork, but has not been noted
in the Garden City formation. Amounts of shale comparable to
that in the Maxfield (about 180 feet, all told) are present only in
the Langston, Ute, and Bloomington formations at Blacksmith
Fork.
Furthermore, flat-pebble conglomerate is abundant throughout
the lower 600 feet (about half) of the Garden City formation at
Blacksmith Fork, but has not been noted in the upper half of the
formation there at all. The relation noted in the Maxfield forma-
tion, of 480 feet of interbedded shales and limestones with flat-
pebble limestone conglomerate in the upper 1o feet only, could
thus not be matched in the Garden City formation. A sequence
almost precisely similar to that of the Maxfield formation does
occur, however, in the Bloomington formation.
The Maxfield formation of the central Wasatch, therefore, is
probably of Cambrian age, and bears a striking likeness to the
Middle Cambrian Bloomington formation of the Bear River
plateau, 75 miles farther north.
The close relation of the Canadian series to the Upper Cambrian
sertes.—If the upper limit of the Cambrian system in northern
Utah has been defined correctly, the changes which took place
between Canadian (or Chazyan) and Trenton time in the Rocky
Mountain-—Great Basin paleogeographic province were more notable
than those which occurred between Upper Cambrian and Canadian
time in that region. The latest sediments assigned to the Upper
Cambrian are of the same type as the Canadian deposits, whereas
the sediments of Trenton age are very different from either of the
former. The erosion preceding the beginning of Trenton sedi-
mentation is known to have been extensive, whereas evidence of
erosion between Upper Cambrian and Canadian time is reported
from only one locality.
The physical evidence thus goes to show that the affinities of
the Western Canadian are rather with the Cambrian than with the
higher Ordovician.
Is the Osarkian system represented here ?—If the Ozarkian period
of Ulrich is represented in the western province, it must be by some
part of the Upper Cambrian—Lower Ordovician sequence above
250 C. W. TOMLINSON
described. In the absence of an adequate description of the fauna
of the typical Ozarkian, it is difficult to make comparison therewith.
The base of the Ordovician was placed by Walcott* at Blacksmith
Fork at the first appearance of cephalopods. As determined by
Richardson’ in the Randolph quadrangle, the base of the Garden
City formation there is marked by the appearance of several genera
of coiled gastropods. The post-Cambrian, pre-Swan Peak (see
below) series thus defined has a thickness on Blacksmith Fork of
1,272 feet, which is 68 per cent greater than the total thickness of
the overlying Ordovician, including the Richmond. As above
noted, Richardson describes a marked unconformity at the top of
the Cambrian in the Randolph area. ‘These faunal distinctions and
this physical evidence may warrant the recognition of the series
in question as a separate system; but there is no evidence yet at
hand to suggest its subdivision into two systems, Ozarkian and
Canadian. It is possible that the St. Charles formation includes a
representative of the former.
Up to the present, however, the Ozarkian has not been recog-
nized in the Rocky Mountains.
THE ORDOVICIAN QUARTZITES AND SANDSTONES
The Eureka and Swan Peak quartzite——In eastern Nevada the
unfossiliferous Eureka quartzite, ranging from 200 to 500 feet in
thickness, lies in apparent conformity upon the Pogonip limestone.
The upper surface of the Eureka quartzite is clearly an irregular
erosion surface. It is overlain by the Lone Mountain limestone,
which carries a Trenton fauna near its base.’
The Eureka quartzite corresponds closely in stratigraphic rela-
tions to the quartzite at Geneva‘ and to the Swan Peak quartzite,
both in northern Utah, which attain a similar thickness. Uncon-
formity is evident above the Swan Peak quartzite in the Randolph
quadrangle,° and farther south the post-Swan Peak erosion locally
tC. D. Walcott, “Cambrian Cordilleran Sections,” Smithsonian Misc. Coll.,
LIII (1910), ror. ;
2 Op. cit., pp. 408-9. 3 Hague, op. cit., pp. 58-59.
4 Eliot Blackwelder, ‘‘New Light on the Geology of the Wasatch Mountains,
Utah,” Bull. Geol. Soc. Amer., XXI (1910), 526-27.
5 Richardson, op. cit., p. 408. 6 Thid.
PALEOZOIC STRATIGRAPHY OF ROCKY MOUNTAINS 251
resulted in the complete removal of the quartzite, so that in the
section exposed on Blacksmith Fork the succeeding dolomites of
Trenton(?) age rest unconformably on the Garden City formation.
.P00%
‘$=:
o None(?)
The Swan Peak quartzite contains a small fauna which has been
referred tentatively to the Chazyan by Ulrich.t Four of the eight
forms identified from the Swan Peak are found also in the similarly
' Op. cit., p. 408.
Fic. 7.—Map showing the extent and thickness of the Chazyan (?) quartzites
252 C. W. TOMLINSON
meager fauna of the quartzite at Geneva. There is no representa-
tive of these quartzites in the section described by Hintze? in Salt
Lake County, Utah, 75 miles south of Blacksmith Fork.
The absence of the Swan Peak quartzite from the Blacksmith
Fork section is especially notable in view of the fact that it occurs
with a thickness of several hundred feet less than 20 miles to the
northeast, in the Randolph area, and also less than 20 miles away
in the opposite direction, near Geneva.
The “Ogden quarisite” in the Uinta Range.—In the south slope
of the Uinta Range a quartzite attaining a maximum thickness
of about 1,100 feet lies between the Mississippian system and the
barren Lodore shales (Cambrian ?). Weeks? correlated this forma-
tion with that which is called in this paper the Swan Peak quartzite,
in the region east of Cache Valley, Utah; and he applied to both
formations the same name, “Ogden quartzite.”’ As no fossils have
been found in the “Ogden quartzite’ in the Uinta Range, the
correlation rests on an insecure basis.
No representative of this quartzite was noted on the north
flank of the Uinta Range.
The sandstone at the base of the Bighorn formation.—The Trenton
dolomites in Wyoming for the most part lie directly on the flat-
pebble conglomerate series, but locally they are accompanied by a
thin basal sandstone carrying a late Black River or early Trenton
fauna, including fish remains. This sandstone is correlated with
the fish-bearing Harding sandstone of Colorado,’ which likewise
is overlain by dolomites of Trenton age. It is probable, as has
usually been considered, that this sandstone member represents an
introductory stage of the Trenton submergence rather than that
it is a deposit of an earlier submergence, separated by an epoch
of erosion from the Trenton proper, as is the case with the Swan
Peak and Eureka quartzites. The faunal lists from the Swan
* Blackwelder, Joc. cit. als. 2 Op. cit.
3 F. B. Weeks, “Stratigraphy and Structure of the Uinta Range,” Bull. Geol. Soc,
Amer., XVIII (1907), 436-37, 441.
4N. H. Darton, op. cit. (1905), p. 551; and Bald Mountain—Dayton Folio, Wyom-
ing, Geol. Atlas U.S., Folio No. 141 (1906), p. 4.
5 N. H. Darton, op. cit. (1905), p. 552; and Folio 141, p. 4.
PALEOZOIC STRATIGRAPHY OF ROCKY MOUNTAINS 253
Peak quartzite’ and the quartzite at Geneva’ include no species,
and only two genera (Orthis and Endoceras) in common with the
published lists of fossils from the Harding sandstone? and the basal
Bighorn sandstone.*
THE MIDDLE AND UPPER ORDOVICIAN DOLOMITES
Extent of the Bighorn dolomite—The Bighorn dolomite was
named by Darton® from its characteristic exposures on both flanks
of the Bighorn Range in northern Wyoming, and by him was cor-
related with the Whitewood limestone of the Black Hills and with
the Fremont limestone of the Front Range of Colorado. The same
author later recognized the Bighorn dolomite in the Owl Creek? and
Wind River’ ranges of Wyoming. Fisher? briefly described its
occurrence in Cedar and Rattlesnake Mountains, west of Cody,
Wyoming. Blackwelder” has identified it in the Gros Ventre and
Teton ranges, farther west.
The Bighorn dolomite in Montana.—Both Darton" and Fisher”
prophesied that the Bighorn dolomite would be found to be included
in the “Jefferson limestone” of Hague, and the truth of this
t Richardson, op. cit., p. 410. 2 Blackwelder, op. cit. (1910), p. 527.
3 N. H. Darton, ‘Fish Remains in Ordovician Rocks in the Bighorn Mountains,
Wyoming, with a Résumé of the Ordovician Geology of the Northwest,’? Bull.
Geol. Soc. Amer., XVII (1905), 563.
4 Tbid., pp. 554-50, footnote.
5 N. H. Darton, ‘‘Comparison of the Stratigraphy of the Black Hills, Bighorn
Mountains, and Rocky Mountain Front Range,” Bull. Geol. Soc. Amer., XV (1904),
379-448.
°N. H. Darton, “Description of the Bald Mountain and Dayton Quadrangles,”
Geol. Atlas U.S., Folio 141 (1906), p. 4.
7N. H. Darton, “‘Geology of the Owl Creek Mountains,” Fifty-ninth Congress,
st session, Senate Document No. 219 (1906), p. 15.
8.N. H. Darton, ‘‘The Paleozoic and Mesozoic of Central Wyoming,” Bull. Geol.
Soc. Amer., XIX (1908), 403-74.
4
9C. A. Fisher, ‘Geology and Water Resources of the Bighorn Basin, Wyoming,”
U.S. Geol. Survey, Prof. Paper 53 (1908), p. 12.
t0 Eliot Blackwelder, unpublished manuscripts, U.S. Geol. Survey.
1 N. H. Darton, ‘‘Fish Remains in Ordovician Rocks in the Bighorn Mountains,
Wyoming, with a Résumé of the Ordovician Geology of the Northwest,” Bull. Geol.
Soc. Amer., XVII (1905), 554.
OWNS Adi
W. TOMLINSON
(Ge.
254
prophecy has now been demonstrated by the writer’s work. The
Bighorn dolomite is characteristically developed throughout the
length of the Absaroka Range, and has yielded Ordovician fossils
ae 6 eee a oe S rr nm eee ee —-—-
—_—-
DUON
(09) 42A0 JON) 3 24ONO
@+29= cei), 96n°¢
SOO pes ==
y)
Za eUONO
Bere — TL
See ee ee ae
a mas ean —
ae
“SUOT}EUIO] Pozeper1I09 pue (UPIOIA
-OpiQ Joddy pure o[ppry]) eWWoLop u10YSIg 94} Jo ssouyDTY) pue yUE}xO oy} BuImoys dep—'e ‘o1,7
_——-—,
ee eee ee,
although this formation is 438 feet
thick in Livingston Peak, it has no known continuation north of
that point, nor west beyond the Gallatin Range.
p)
Se
e
in the Crandall (Wyoming) and Livingston (Montana) quadrangles.
It is a remarkable fact that
PALEOZOIC STRATIGRAPHY OF ROCKY MOUNTAINS 255
Members 2-4.—There can be little doubt as to the validity of
most of the correlations of Trenton strata shown in the preceding
table and diagrams, except in Utah and Nevada. The main mas-
sive member (No. 4) of the Lower Bighorn, with its accompanying
basal weaker strata (Members 2 and 3) and its cap of smooth,
nearly chalky, dolomite, constitutes a highly characteristic and
almost unique series. Furthermore, Member 2 is fossiliferous,
and Member 4 is sparingly so in nearly all localities.
The upper half of the massive member in the Blacksmith Fork
section, which the writer has called the Lower Fish Haven dolomite,
is similar in all essential respects to the massive Trenton member
of the typical Bighorn. It is marked off from the Richmond above
by a conglomerate, and disconformity at its base is sufficiently |
indicated by the absence of the (Chazyan ?) quartzite which inter-
venes at that horizon elsewhere in northern Utah. This lower
Fish Haven dolomite carries Halysites, which is not known from
rocks older than Mohawkian.
The lower part of the Lone Mountain limestone, which uncon-
formably overlies the Eureka quartzite in western Nevada, carries
a fauna assigned by Walcott* to the Trenton. Ulrich? has voiced
the opinion that part of the Lone Mountain limestone is older than
the Bighorn dolomite; but it is probable that the former formation
contains a representative of the Trenton series.
Members 5-7; the Leigh formation—Members 5-7 of the Middle
and Upper Ordovician series constitute a distinct and very widely
developed unit. In the Goose Creek Ridge section there is little
ground for differentiating these three from each other; but the
lowest Richmond fauna occurs in the beds there marked as Member
6. In the Crandall Creek and Dead Indian Creek sections there is
a conspicuous surface of disconformity, with a basal breccia, at
the base of Member 6. As this is the only disconformity for which
physical evidence has so far been noted anywhere within the limits
of the Bighorn formation, and as it coincides (by lithologic correla-
t Arnold Hague, op. cif., pp. 61-62, 196-97; also appendix by C. D. Walcott,
PP. 324-25.
2 Cf. Bailey Willis, ‘‘Index to the Stratigraphy of North America,’ U.S. Geol.
Survey, Prof. Paper 71 (1912), p. 169.
256 C. W. TOMLINSON
tion of beds both above and below) with the base of the known
Richmond part of the Bighorn in the type locality, it probably
represents the hiatus which was inferred by Darton between the
Trenton and Richmond members of the Bighorn.
In the Teton Range, there is likewise an unconformity at the
base of Member 6. Blackwelder’ proposes the recognition of
Members 6 and 7 in that region, and of corresponding strata in the
Gros Ventre Range, as a distinct formation, to be called the Lezgh,
from its typical development on Leigh Creek, in the Teton Range.
In view of the fact that this group of strata, in its type locality,
is bounded both above and below by unconformity, and is lithologi-
cally quite distinct from the underlying massive member (Member
5 is not present in the Teton River section), its recognition as a
separate formation seems justified. In the Absaroka Range the
corresponding beds differ little in character from the upper part of
the Trenton series (Member 5), but are marked off from it by uncon-
formity, as just described.
Between Members 4 and 8 of the Livingston Peak section, light-
colored dolomites of the Leigh type occur interbedded with darker,
gray-brown, more coarsely crystalline dolomites. In the Black-
smith Fork section there is an interbedding of light and dark
strata through a thickness of 130 feet above the conglomeratic
horizon, which is taken as the probable base of the Richmond series.
Above this sequence there is an 8-foot stratum corresponding
closely to the typical Leigh in character, and directly underlying
the main massive part of the Fish Haven, which is correlated with
Member 8. The first sediments deposited after the pre-Richmond
emergence seem to be more variable in character from place to
place than are the strata above them, or the members of the Trenton
series.
Members & and g.—Member 8 is in some places, in lithologic
characters, essentially similar to Member 4, but is in no case quite
so thick as the latter. On Goose Creek Ridge there are only two
12-foot massive beds of this type, themselves separated by 20 feet
of less resistant dolomites, between the typical Leigh (Members
6 and 7) below and the main fossiliferous, thin-bedded part of the
tEliot Blackwelder, personal note.
PALEOZOIC STRATIGRAPHY OF ROCKY MOUNTAINS 257
Richmond series above. It is possible that the much thicker
massive dolomite characterizing the Upper Bighorn in the Absaroka
Range is equivalent to a part of this thin-bedded series (called
Member g), as well as to the underlying more massive beds. On
Blacksmith Fork, Member 8 forms the main body of the Upper
Fish Haven dolomite. In common with several other parts of the
Ordovician system, it is somewhat darker in color there than in
Wyoming.
The highly fossiliferous, thin-bedded dolomites (Member 9) of
the Upper Bighorn in the Bighorn Range unfortunately are not
typically developed elsewhere.
There is no representative of the Bighorn in Hintze’s' section
in the central Wasatch, nor in the Uinta Range.’
1F, F. Hintze, Jr., op. cit. 27F. B. Weeks, op. cit.
A FEW INTERESTING PHENOMENA ON THE ERUPTION
OF USU
Y. OINOUYE
Imperial Tohoku University, Sapporo, Japan
The eruptions of Japanese volcanoes for the past fifty years
have been almost invariably of the Strombolian type. But recently
there have been displayed five different types which may be listed
as follows: (1) an appearance of a new volcanic island,’ (2) a new
lava dome in the crater of Tarumai, (3) 45 craterlets on the slope
of Mount Usu, (4) ejecting lava up in the craters of Asama’? and
Mihara,} and (5) lava flows on Sakurajima. Partial descriptions
of the Tarumai and Usu have been published by the writer,’ while
an account by Professor B. Koto of the third will appear in the
near future.
In the southern part of Hokkaido, in North Japan, three vol-
canic eruptions took place between 1905 and 1910. Komagatake
was in eruption in August, 1905, Tarumai in April, 1909, and Usu
in July, toro. A line connecting these three volcanoes lies in a
northeast to southwest direction and represents the northern
extremity of the Nasu volcanic chain. The three volcanoes men-
tioned are about equally distant from each other (48 km.). The
explosion of Komagatake was simple and on a small scale, ejecting
fragments around the crater and ashes around the foot of the
volcano for a few days only, while that of the Tarumai’ was more
=T. Wakimidzut, “Report on the Ephemeral Volcanic Island in the Iwojima
Group,” Bulletin of the Imperial Earthquake Investigation Committee, No. 56 (1907).
2F. Omori, Bulletin of the Imperial Earthquake Investigation Committee, VI, No. 1
(1912).
3 Y. Okamura, Bulletin of the Imperial Geological Survey of Japan, No. 48 (1914).
4 Report of the Imperial Earthquake Investigation Committee, No. 64 (1909). Offi-
cial Report of Hokkaido Colonization (1910).
5 Y. Oinouye: Report of the Imperial Earthquake Investigation Committee, No. 64
(1911); H. Shimotomai, Zeitschrift der Gesellschaft fiir Erdkunde zu Berlin, No. 9
(1912).
258
INTERESTING PHENOMENA ON THE ERUPTION OF USU 259
severe. The Usu eruption is the latest one among the three,
which was quite similar to that on Etna in September, sort.
This volcano is located between the other two, in longitude
E. 140° 40’ 30” and latitude N. 42° 33’, and lies between ‘‘ Volcano
Bay”’ on the south and Lake Toya (80 m. higher than sea-level)
on the north. Usu is a low, conical, active volcano, 736 m. above
the sea, and has a crater 2 km. in diameter, within which there are
Nishi-Meruyama
LAKE TONA
Tokotan
= Mud Flow
(i Elevated Portion MS
© Craterlet
i
a ae M8 Nishikohan
7 Higashi-Maruyama
e Village
nN Fault
Abutg
eB opuyarsc?
400%
BAY ;
OF Cs
VOLCANO
Usu
AY OF US)
Fic. 1.—Map of Volcano Usu, showing craterlets on the northern slope
two domes occupying respectively the east and the west end of the
crater (Fig. 1). The eastern of these domes, O-usu (736 m. AT),
looks new, while the western one, Ko-usu (609 m. AT), appears
much older. Ko-usu has a few small, steaming pits on the top of
its dome, while O-usu has one only on the west side of its dome.
The topography, geology, and history of Usu have been well
described by Professors F. Omori' and D. Sato.2. Hence only the
especially interesting details of the eruptions will be discussed
here.
* Bulletin of the Imperial Earthquake Investigation Committee, V, No. 1 (1911).
2 Bulletin of the Imperial Geological Survey of Japan, XXIII, No. 1 (1913).
260 Y. OINOUYE
I. EARTHQUAKES AND ROARINGS
Preceding the eruption there were frequent earthquakes, seem-
ing to repeat the past history of the mountain, which has always
exhibited the ‘‘foreshocks”’ in advance of an eruption. But the
writer believes that the occurrence of so many earthquakes in the
neighborhood of a volcano within the limits of Japan is a rare
phenomenon. It was rumored that slight earth movements were
Fic. 2.—A fissure on the road near Abuta
noticed six days before the eruption. But, as observed by a few
persons, the first earthquake began on the evening of July 21, four
days in advance of the eruption, and successive earth tremblings
were felt from the morning of July 22, continuing through the
eruption and for two months thereafter. Numbers of these
quakes were felt at Nishimombetsu, 8.4 km. southeast of Usu,
25 on July 22, 110 on July 23, 354 on July 24, 163 on July 25, and
thereafter in gradually decreasing numbers. It was on the evening
of July 25 that the first eruption took place, and, after it had relieved
the strains to some extent, the quakes began to decrease in number.
INTERESTING PHENOMENA ON THE ERUPTION OF USU 261
Fic. 4.—A monument shaken down by the severe earthquake at 4:30 P.M. on
July 24, 1910.
262 Y. OINOUVE
The report of the Municipal Office and the members of the Meteoro-
logical Observatory of Sapporo and Hakodate give the numbers in
Table I.
TABLE I
NUMBER OF EARTHQUAKES OBSERVED AT NISHIMOMBETSU
Date Violent Strong Weak Tremor Total
SJsye NN oe OR ME UM tec Da UAROH Ca 13 12 25
Ae eR (ae Say eta 8 48 54 mite)
2 Ale I 28 134 150 313+40.5
25s I 19 58 85 163
PAS TNSGUE AIS ae SL Nr I It 16 28
Xs Ee Slee aie aia 3 14 5 22
Po AE ait A Berne seas dle 6 3 8 17
PAG Rihaial Maes inet eae I 2 9 12
FO eh eo liallinante e kt I 3 I &K
Te Ta A MU MEN Re RUN Tee SRN age Me teiea 2 I 3
PANU USE sae ceili econo shepe asec etn alcage eeepc recur le 2 4 6
PD ap ie aersisera ap ctr cecerag [nse UCU cin ta UTE Dy er Ge 2 2
* Lack of observation for three hours. The number is estimated by means of an average for the
three preceding and the three following hours.
From hourly observations the following results were obtained.
From July 22, 7:00 A.M., to July 23, 7:00P.M., 36 hrs., 66 quakes, 1.8 per hr.
From July 23, 7:00 P.M., to July 25, 8:00 A.M., 37 hrs., 533 quakes, 14.4 per hr.
From July 25, 8:00 a.M., to July 25, 10:00 P.M., 14 hrs., 48 quakes, 3.4 per hr.
The writer’s visit to Mount Usu was made on the afternoon
of July 24, amid the climax of the quaking. At that time the quakes
occurred rather oftener than once in five minutes. The houses
trembled so from the subterranean violence that the windows
rattled continually throughout the entire day, and made so much
noise that no one could stay within the houses. It was noticed
that every quake was preceded by the sound which seemed to
come from deep within the earth, or as if heavy artillery were being
fired in the distance. But sometimes on the east side of the moun- |
tain, or in the direction of Volcano Bay, probably owing to the
echoes, the same sound was heard. It frequently happened that
the sound was first heard in the distance; then a landslide was
seen on the dome of O-usu; and following almost immediately
the quivering of the earth was felt. A year previous, when the
writer visited Usu, a small column of steam was seen to rise from
the small pit on the west side of O-usu, and this was the same in
INTERESTING PHENOMENA ON THE ERUPTION OF USU
263
Fic. 5.—The largest mud cone at Usu village. Taken July 30, 1910
Fic. 6.—Numerous mud cones in the Bay of Usu.
Taken July 30, 1910
264 Y. OINOUYE
Fic. 7.—A fault of 1 m. throw, at the west foot of the Kompirayama. Taken
August 2, 1910.
Frc. 8.—The same fault which has increased its throw to 2 m. Taken Sep-
tember 4, IgIo.
INTERESTING PHENOMENA ON THE ERUPTION OF USU 265
amount when the second visit was made during the time of the
eruption under discussion. The surrounding country was noted
Fic. 9.—The first explosion crater on the Kompirayama. Taken July 26, 1910
to be the same topographically as it had been on the previous visit.
Judging from the history of this volcano, the writer recognized
Y. OINOUYE
LAKE TOYA
LAKE TOYA
Fic. roa.—A sketch of craters from the west. 5:00 P.M., July 27, 1910
Fic. 10b.—A sketch of craters from the northwest. 11:20 P.M., July 28, 1910
Fic. roc.—A sketch of craters from the north. 6:00 P.M., July 29, 1910
INTERESTING PHENOMENA ON THE ERUPTION OF USU_ 267
that the preliminary warnings were the symptoms of an eruption,
and, watching the crater every moment during the day and night
and the following day, he observed in detail the phenomena. The
number of earthquakes, as well as their vigor and intensity, in-
creased. No one could indulge in sound sleep in the neighborhood
of the volcano. Cannonading and trembling developed till the
introductory explosion took place. Among the several hundred
Fic. 11.—A great column of smoke at the top of Nishimaruyama beyond the
steaming mud flow. Taken August 2, 1910.
earthquakes two violent ones are worthy of mention, one at
4330 P.M. on July 24, and the other at 5:00 P.M. on July 25. Monu-
ments fell, houses were badly damaged, the earth was ruptured,
and many mud cones were formed around the volcano (Figs. 2-6).
The earthquake wave reached an average radial distance of 65 km.
outward from Usu, except in the southwesterly direction, where
it reached 140 km. The earth’s shaking abruptly decreased after
the first explosion, suggesting that strains which had been accumu-
lating were then relieved.
268 Y. OINOUVE
WI. EFFECT OF THE EARTHQUAKES
There were several phenomena of interest due to the preliminary
earthquakes, such as fissures, faults, and the building of mud
cones.
1. Fissures—Many ruptures were made within the circle of
severe Shaking of the earth, especially on the west side of the
Fic. 12.—South scarp of ‘‘graben”’ at the top of Kompirayama. Taken July 14,
IQII.
volcano. The directions of the fissures were almost parallel to
the coast line, i.e., northwest to southeast, and their width was
from 3 cm. to 40 cm. (Fig. 2). Close to the mountain, on the
west side, the direction changed to east-west.
2. Faults—Two distinct faults extending east and west were
made on the west foot of the Usu. Stepping down toward the
north (downthrow side on the north), the throw of the southern
fault measured 30 cm. and that of the northern one 1 m. The
former extended about 50 m. and the latter 600 m. in length. On
September 2 an additional throw of 1 m. was noticed, developing
INTERESTING PHENOMENA ON THE ERUPTION OF USU_ 269
numerous small parallel fault fractures besides showing 2 m. of
horizontal shifting (Figs. 7 and 8).
3. Mud cones.——The mud cones are small mounds of mud and
sand, well stratified and laminated. They range in size from.
several centimeters to three meters in diameter, and are flat and
conical in shape, the angle of slope being from 3 to 16 degrees.
The smallest cone seen measured 10 cm. in diameter, and the largest
Fic. 13.—North scarp of ‘‘graben’’ at the top of Kompirayama.: Taken July 14,
IgIt.
5m. (Fig. 5). The height of the former was 3 cm. and that of the
latter 60 cm. Great numbers of such cones were formed in the
bay at the southwestern foot of the mountain, distributed irregu-
larly upon the tidal flat (Fig. 6). About 200 m. from the shore
line there is a row of such cones trending generally northwest-
southeast. While this row of cones is roughly parallel to the shore
line, and consequently fairly straight, there are numerous bends in
the line. From the structure of the cones it follows that there must
have been a periodical eruption of the sand and mud. The lamina-
tions, ranging from 5 mm. up to 3 or 4.cm., are roughly proportional
270 Y. OINOUYE
to the size of the cone, the thicker laminae being found in the larger
cones. As is usual with all the cones of this region, it was cold
water that issued from them before the eruption of the volcano.
On the north side of Usu a few cones were found on the flat farm
land at the foot of the mountain. At no other place in the neigh-
borhood were these phenomena observed. All the cones were
Fic. 14.—Step fault at the west foot of Kompirayama. Taken August 18, 1910
formed by the first severe earthquake, which occurred at 4:30 P.M.
on July 24, 1910. The phenomenon is not a peculiar one, for such
cones have been reported at many places where strong earthquakes
have taken place. They are invariably located along the crack
formed by the earthquake where the ground-water issuing through
the newly opened vent brings sand and mud with it to the surface.
After the eruption of the volcano the mud cones ceased to be
active and were gradually obliterated by the process of erosion.
4. Rise of the water-level in near-by wells.—Practically all the
wells in the neighborhood of the volcano showed a rise in the
INTERESTING PHENOMENA ON THE ERUPTION OF USU 271
water-level; very few showed a decrease. In most cases it was
noted that the water in the wells increased to about double the
normal volume, while at the same time it became turbid and dirty,
owing to the particles of dry mud which fell from the wells into the
water below. The rivers of the region also became brown and
turbid from slumping of the clay banks. On the southeast side of
the mountain several new springs were formed which are still flowing.
TII. EXPLOSIONS
After July 22 fre-
quent earthquakes took
place, their intensity and
numbers increasing hour
by hour till 10:00 P.M.
on July 25, when the
first explosion took place
on the northwest side of
Kompirayama, a para-
sitic cone, on the north-
: Fic. 15.—T wo groups of craterlets. Group I is
west slope of the main ocated at the southwest of Group II.
volcano (Fig.9). Fora
few hours red-hot bombs were ejected on the north side of the
cone and made numberless holes in the roofs of the houses in
the near vicinity. When the writer visited the region on the
following morning, the vent was entirely free from escaping
steam, making it possible for him to descend into the crater.
At 2:13 P.M. on July 26, a second explosion, preceded by roaring
and trembling, took place 200 m. southeast of the first crater
accompanying two small explosions. This explosion ejected
black and white smoke to a height of about 700 m. That night
the smoke stopped for a few hours, but again, beginning at three
o’clock in the morning, the roaring became louder and louder as of
strong thunder near by, till four o’clock, when it gradually sub-
sided. Meanwhile frequent earthquakes accompanied the forma-
tion of three or four explosion craters. In the afternoon of the
next day, July 27, the writer saw the ejection of smoke in two
craters east of the crater mentioned above (Fig. toa). At 7:00 A.M.
272 Y. OINOUVE
on July 28, roaring again began, and two explosions took place in
sight of the writer at 11:20 A.M. (Fig. 100). The loud roaring
continued till eight o’clock in the evening. On the same day there
was heavy rain all day accompanied by loud thunder, intense
lightning, loud roaring in the ground, and much dense smoke
hiding the mountain entirely from view. Before 9:00 A.M. on
July 29, judging from the numbers of smoke vents found on that
morning, several more explosion craters were formed (Fig. toc).
i)
Fic. 16.—A group of craterlets on the second group. Taken September 4, 1910
On August 2 two new craters were formed at the top of Nishi-
maruyama, a parasitic cone. For a week these two craters poured
forth an astonishing amount of smoke in a large black column such
as was not seen from any other cone (Fig. 11). The volume
suddenly decreased at the end of one week. Thus in the period
of greatest activity, from July 25 till August 2, the number of
craters formed amounted to 15. Thereafter, on August 7 and 8,
on September 3, and on October 2, small explosions occurred, so
that the total number of craters reached 45. Besides these craters
numerous crevices and faults developed on the side of the moun-
INTERESTING PHENOMENA ON THE ERUPTION OF USU 273
tain. At the top of the small parasitic cone, Kompirayama, a
‘““graben’’ was formed about 30 m. in maximum depth, 100 m.
wide, and roughly 500 m. in length (Figs. 12, 13). Westward,
across the road, the same faulting was found to be “‘step faulting,”
with the northern blocks downthrown (Fig. 14). It seems appar-
ent, therefore, that the northern fault of the “‘graben”’ is a “‘scissor
fault,” reversing its throw on crossing the road. Close to the
second group" of craterlets, at the same time, were formed many
Fic. 17.—Mud-flow from crater No. VII. Taken August 2, 1910
faults, among which were some forming fault scarps of 5-7 m. in
height. The trend of both systems of faults is west-northwest by
east-southeast. It is noteworthy that such a large number oi
craterlets formed in the ten weeks of the eruption.
t. Process of explosion—The order of formation of these
craterlets was as follows: In the beginning a cannon-like sound
was heard, the ground cracked open in a straight line in the form
of a V-shaped crack (Fig. 9), and white smoke issued from the
vent. Then followed black smoke together with sand and ashes.
t See Fig. 15;
274 ¥. OINOUVE
This is the normal order of eruptions for all the craters. The ashes
and bombs ejected from the fissures upon the sides of the crevice,
after a day, or at most a few weeks, built upacone. The rapidity
with which the cones were built depended on the size of the orifice
and the amount of ejectamenta. The angle of slope of the cones
ranged from 15 to 30 degrees, the steeper ones being made
of the rather coarse material. As the cones grew, the old vents
became quiescent and new ones broke out.on the side of the
Fic. 18.— Puff cones” on the mud-flow. Taken August 19, 1911
cone, and the ejectamenta, filling the old crater, sometimes
obliterated it.
The life-history of the different cones was not the same, some
of them becoming quiescent after only one explosion, others con-
tinuing for several days to roar and emit black smoke and to
build cones. Some of them-emitted smoke intermittently, being
active for a few days and then quiet for a few days and then active
again, etc. Besides the smoke, ashes, sand, and bombs, a large
amount of mud’ flowed from five of the craters.
t Jour. Geol., XXIV, No. 6.
INTERESTING PHENOMENA ON THE ERUPTION OF USU 275
2. Explosion craters—The 45 craters are arranged in two
groups (Fig. 15). The first, consisting of 16 craters lying north-
northwest of Mount Usu, is aligned in a west-northwest to east-
southeast direction. The second group, lying 800 m. northeast
of the former, is composed of 29 cones lying in a line parallel to the
first group (Fig. 16). The altitude, date of formation, size, shape,
and the life-history etc., of each cone in the two groups are tabu-
lated in Table IT.
Fic. 19.—Ejection of smoke from crater No. XLII. Taken August 3, 1910
The formation of the cones apparently has no order, though a
few in the first of the groups described above formed in sequence,
starting from the northwestern end and proceeding toward the east.
But in general the action was begun in the first group and finished
in the second group.
3. Ejectamenta.—Ash, sand, and bombs, with SO, and H.S,
were ejected in the black smoke and carried to the lee by the wind,
some of the ashes quite a distance. From the first explosion ashes
were carried 44 km. toward the northwest of the Usu. Later, ashes
from a later eruption were carried 4 km. toward the east and 20 km.
Y. OINOUYVE
276
yqnos suado 193¥vI-
QAT}OV YSOUI 94}
pue smop-pnur ‘zId ysazie] oY} BYCU PUL 9}TUN S19}eID JBI],
ysva preMo} suado 3903219
TITAX XX Jo ByuaUIe}Af9 Aq poddeysaAo ATW ed pue smog-pnyy
equowleyal[a Aq
way} sdevjiaao pur ‘TAX XX pue AKXX wey} sare 10
Surures}s [[IJs pue ‘6z A[nf uo smop-pnyy
J9}e1D 9Y} Ul [[e@M YIIOU aq} UO ZuIUIeI}S
€1 Jsn3ny uo 9u0d ay} saja;duioo pue ‘eyuowuejzsefa YOnL
Jaquisydasg jo SUIUUIZaq 9} UI 9AI}Ie YSOJ[
DAT} JSOUL OY} PU 19}eID YSIZILT PUODES 9Y J,
Jaquiaydag jo SUIUUTSaq 9} UI dATIOV YSOPT
Sururesys [[NS
YoaM & IOJ 9YOUIS Yor]G sq
YOIM V IOJ 9YOUIS YORIG Sig
YoaM IO] Yous Yoel Sig
Aqrquenb e013 UI SMOY-PpNUl pue ySsaM PIeMO} SUadO I9zeIZ)
4saMm preMo}y suado 19}8I_
ysnsny jo Suruursoq [IWUN A[Iep ay¥ours Yoeyq srw
SMOP-PNIY
‘Ul Ooo'T 03 dn YZIY VYOUIS
SIMOY Mj B 10} squlog pure sayse s}afa puke 9YOUWS sanssy
1161 ‘Avy [UN Furures}s
SHUVINAY
° ° 6
° ° L
° ° Ir
of os 06
ord or of
° ° or
fe} ° II
° ° Le
oz or gI
144 ford II
loRomoM cM om odode meMememe mcm eh olmeome Mom em eM elec)
iP) an
gyro eooeooONd00D0D0DDDtADDDODIO AOC
Lal ee
ve)
H
adoys 481eH|yIdoq
aANO)
o1Xoz asdiy[q SSr =
zz punoy oor ro) ¢
$tzxXoLl asda oOLl1 or
oStXoIz | asd Oz oI
ogXoL asda Oz%z 9
oF X 69 asda SLY 9
ov X79 asdiy[a Sor 4
69 punoy OO Teas see
zZXvS asdi[q OO Tear |e
oLX Sg asdiqyq Clef = tee
oz XoL asda OO Cea pce cae
re punoy o0z :
to punoy Soz ¢
gL punoy Chi |B areee
igs punoy OS ees eeaas
06 punoy OC Cee |eae
Six Lz asdiy[q OST areal beens
g&xos asdiyy a AB Ree
LY XoS asdi[q ose L
¥g punoy osz 8
of! punoy yd v1
£0 punoy CUE | ee eee
zvVXzg asda oz 8
81X69 asd q Ol Za |e
gIXor asd] aq ogt II
£z punoy oor II
gI punoy ogI Ir
oor punoy og! L
gf punoy ogI ZI
Lz punoy oz v
QI punoy DISA =| Pees 2
Sv punoy Se |eeenes
16 punoy obz 6
St punoy OO TR
gi punoy OC Ogee |=
oI punoy lorad :
ZzXK OL asdiy ay OC. Cjrmeen ieee
ze punoy Soz $s
feXx2P asdiyy ay, SLz z
vz punoy Sle z
£1X16 asdiy[a ove I
of 9£ asdiy[q oLlz z
1¢ punoy osz €1
SI punoy Sbz €1
SI punoy Szz £1
UL uw
Jayomeiqg, | adeys
GGALIL1Ty |wACIC,
UALVAD
‘Wy 00:6
‘WV 00:6
"WV 00:6
‘WV O7t:1II
‘WV OZ:II
‘WV OO:f
avd
“°° AVIX
ts TTTARKX
mB TIAKXX
‘ON
Tl aTa&vL
INTERESTING PHENOMENA ON THE ERUPTION OF USU_ 277
toward the south, the amount being greatest on the northwest side
of the mountain. In the Kompirayama region the ashes that fell
formed a layer up to 8 and tocm. in thickness, while at the distance
of 1km.from the mountain thicknesses of 3 mm. to 1 cm. were found.
On the north side of the second group of cones, a general thickness
of 1 cm. was found, while at their very foot the layer was 30 cm.
in thickness. But on the east side and south side of the mountain
Frc. 20.—Bombs and mud-flow from crater No. XLIT. Taken July 31, 1910
very little ash was found. Besides the ash, sand, and bombs, from
five of the craters mud and hot water were ejected. Among these
five craters No. VIII was the first to erupt (Fig. 17), while No. XIII
ejected the largest quantity of mud (Fig. 11). From the craters
to the lake is an expanse of mud which flowed out to a width of
200 m., a length of 500 m., and a thickness of 1.5 m. In addition
to this great quantity of mud on the land there was a large amount
that flowed into the lake. The mud is composed of fine, gray-
colored plagioclase, hypersthene, augite, and magnetite, with a
278 Y. OINOUVE
small amount of hematite, together with glass in an amount com-
parable with that of the feldspar. Cone No. XVIII ejected the mud
periodically in a geyser-like fashion. The mud contained a large
amount of gas which came to the surface of the mud-flow after it had
almost solidified, making “puff cones’’*' in great numbers (Fig. 18).
The materials of the mud-flows, the sand of the seashore, and
the substance in the mud cones mentioned above, when compared
Fic. 21.—Heavily burdened trees near the craters. Taken October 16, 1910
under the microscope, were found to be identical in composition.
The fineness, however, is variable; the size of grains in the beach
sand being the largest, that in the mud cones intermediate, and
that in the mud-flow the finest. The base of the volcano Usu is
composed of brown pumice, uniform in constitution throughout
the whole region, and the fact that the mud-flows, the cones, and
the sand of the beach are alike in composition suggests that they all
came from some common source, which in all probability les
horizontally and extends not much below the level of the sea.
t Jour. Geol., XXIV, No. 6.
INTERESTING PHENOMENA ON THE ERUPTION OF USU 279
From craters Nos. XXV and XLII great quantities of bombs and
sand were intermittently ejected to a height of 700 m. (Figs. 19, 20).
The ejected bombs and ashes frequently took the shape of serrate
peaks and pinnacles which rose alternately to great heights and
then sank back’as another one shot up. It was noticed that
descending bombs, when struck by rising ones, produced loud
reports like the explosion of firecrackers. White, comet-like
tails followed the bombs into the air. The largest bomb measured
sate
Fic. 22.—Houses inclined 12° owing to the elevation of the left-hand side. Notice
a man standing straight. Taken September 4, Io1o.
was 25 cm. in diameter and was of the characteristic irregular
and rounded shape. The largest hole noted, formed by a falling
bomb, was 3 m. in diameter by 2 m. in depth.
Petrography of the bombs (augite-hypersthene-andesite).—The
bombs ejected from the several craterlets are quite similar, though
the percentages of the constituent minerals are slightly different.
A brief description follows:
Megascopically, the bombs are dark gray, porous, roundish in
shape, and less than 25 cm. across. The pores are very abundant
at the surface, slightly less numerous within, and range in diameter
280 Y. OINOUYE
from 1 mm. to 2 cm. The majority of the pores are filled with
ashes and sand. Phenocrysts of white plagioclase, which do not
exceed 3 mm. in size, produce a porphyritic texture. There are also
crystals of dark-colored pyroxene, but they are small and not
abundant.
Microscopically, the rock has a hyalopilitic-porphyritic texture.
The groundmass consists of dark-brown glass with minute crystals
Fic. 23.—New elevated mountain seen from the east. Taken December 23, tg10
of plagioclase, augite, and hypersthene, and the phenocrysts of
colorless plagioclase and green pyroxenes.
Plagioclase is the chief constituent. It is either tabular or
equidimensional, and polysynthetic twins and zonal structures are
remarkably well developed. Zonally or irregularly included in some
of the crystals are patches of brown glass and minute grains of
pyroxene.
The extinction angle on the M face of the plagioclase is between
— 28° and —30°, and the maximum extinction angle in the sym-
metrical zone is +32°, showing it to be labradorite.
The hypersthene has a slender prismatic habit and strong
pleochroism, green to reddish brown.
INTERESTING PHENOMENA ON THE ERUPTION OF USU
Augite is usually small in size, mostly in the groundmass.
The rock may be formulated as follows:
Lab., labradorite.
Be
Lab..3+Hyp.s+Aug.,
Goo Glassy+Lab.1s+Hyp.;+Aug.,
Py, 40 per cent of phenocryst.
Geo, 60 per cent of groundmass.
Hyp., hypersthene.
Aug., augite.
281
Chemical composition.—The rock is rather basic, low in SiO.,
The
writer found a great similarity in mineralogical and chemical
composition between the bomb and the lava which forms the old
crater ring of the main volcano, as shown by Table ITI, which gives
an analysis of the bombs, with other similar rocks for reference.
high in Al,O,, CaO, and iron, so that some might call it basalt.
TABLE III
I Il Ill IV Vv VI Vil Vill
SIO ee enna 52.40.| 51.86 | 52.88 | 51.32 | 50.16 | 52.02 | 52.86 | 51.12
ENO aus aiy Ane D750) 20.100)|) 20: 53,| 17.84 | 17.07 | 17.24 | 28.25 | 19.50
He LOR es eyasten ls Saisie 4.46 QeASalaAe sa 2.23 7.90 | 6.61 2.86
HeO i Mie hole TROT SS One Ons On Oh OU COm 2s 53) le snag I 10.53
IAM Ifk OV ae etoreaieleg BE73 2.87 QrOSM AS TOM tAgrOUle ama iiln LAw oT. Tana
MnO Me okie oe OV TOM On 2 OO; DON ee tia 0.30 tre 0.16 | 0.65
CaQen ae ee 230) 10.37) |) T1209 ONS Te EE SS rr AG7 9.58 | 90.54
Re Oe eae hiies a2 They f ally) Iectoyss |] ab FO) || anaes 2 SOM HOHOON O00) | 1ON57
INasOn aii OSU us TO2H satan sTOla ls 3 so) 203810 saan Ni ig lnt
NEO Jape aregiet aa 0.57 On20N |) (Onl TAOS ies pcesios 0725) |b 0.. 00) ) (Ont
ULO Ne ea TROON Riarnc eee eosin: eps poe cual este eal Reese) eli giteatcc eso 0.86
1224 OSA Sue Ara eae COPS ANH Ir Weather sieiat anita svar RUM cMainye Dena eat, ducts cae ech OMAN toy On
I. A bomb from Usu, analyzed by Bulletin of the Imperial Geological Survey of Japan, XXIII, No. r.
Sapporo.
V. Augite-andesite, Kilauea, Hawaii.
. Basalt ? Yate Volcano, Patagonia.
. Pyroxene-andesite, Choa-shen, Kamchatka.
. Basalt, Goentoer lava, Java.
III to VIII taken from J. P. Iddings, Igneous Rocks.
IV.
DAMAGE
. Lava of old crater ring on Mount Usu, analyzed in the same laboratory.
. Luciite, Luciberg, Odenwald Hesse.
. Mean value of three bombs, analyzed in the laboratory of geology in the Agricultural College,
By the fracturing of the earth and the explosions, the deep,
beautiful forest on the slopes of the mountain was destroyed.
The leaves were all stripped from the trees, the greater number of
282 Y. OINOUYE
which were broken and shattered. Many were blown out of the
ground by the explosion, while others were buried and broken by
the fall of bombs and ashes (Fig. 21). The bombs, however, were
not thrown more than 500 m. from the craters, but the sand and
ashes were driven to a distance of several kilometers. Often
heavy showers of sand and ashes were seen to fall in localized
areas, in many places forming long strips of débris on the land.
At one place in a field of barley the strip measured 3 m. wide and
was traced for a distance of 200 m. While in the air these masses
of ejectamenta looked like a jet of water issuing from a hose.
This effect was produced by air currents concentrating the material
into long lines. The damage done by falling ash, including injury
to farm land as well as destruction of houses, etc., was heavy
within a radius of 2 km. of the craters. The most severe damage
by ash and mud-flow amounted to 3 sq. km. of land covered. Five
houses were carried down to the lake by the mud-flow, and a few
houses were buried by the heavy ashes, while five other houses were
shattered as a result of the local undulation of the land. At Abuta,
a distance of 4 km. from the nearest crater toward the northwest,
a monument and a small house fell, together with three brick walls
(Figs. 1, 3, 4) and two plaster ones of a storehouse. In the same
village many cracks developed in the walls of the houses.
V. CHANGE OF TOPOGRAPHY
On July 28 the writer found the rise of the water of Lake Toya
on the north side to be about 30 cm. On August 6 Dr. Omori:
found a lowering of the water-level on the south side of the lake.
From August 20 it was noticed that the north side of the second
group of craters began to rise. This elevation (155 m. high from
the lake-level, according to Professor F. Omori) continued till the
end of November. ‘The slope of the southern shore of the lake was
about 5 degrees. It then gradually rose to a slope of 30 degrees at
the top of the elevation, and 22 degrees on its flank. A photograph
taken by the writer on September 4 shows a house which, originally
constructed upright, was then inclined 12 degrees from the vertical.
Two days later this house had collapsed (Fig. 22). Before the
INTERESTING PHENOMENA ON THE ERUPTION OF USU_ 283
eruption the cone Nishimaruyama could be seen from the village
of Nishikohan, but as a result of the elevation which took place
the view was obstructed. The area of elevated land is about 2 km.
in length by 1 km. in width to the edge of the lake, and, judging
from soundings made, it extends another kilometer under the
water (Fig. 23). The maximum height of the elevation measured
about 120 m. (Figs. 24, 25, 26).7
Fic. 24. Mountain slope in the beginning of eruption. Taken July 29, 1910
Mr. Ito, of the Sapporo Meteorological Observatory, found a
lowering of 36 m. on the top of the new mountain in April, rorr,
while Mr. lizuka, of the Imperial Geological Survey, recorded 43 m.
lowering in July, tg11, by an aneroid barometer.
When the gases involved in the lava are expelled in a great
quantity, a decrease of volume will take place, and the lowering
of the mountain should result from this shrinkage.
1 This measurement was made by comparing graphically and to scale the photo-
graph taken before the elevation with that taken afterward. This checked well with
the reading of the aneroid barometer which nearly coincides with the map of the
Imperial Geological Survey of Japan.
284 Y. OINOUYE
Furthermore, there is a remarkable change of height in Usu
proper, as is shown by the map of the Imperial Geological Survey,
July, to11. In the topographical map published by Hokkaidocho,
the height of O-usu is recorded as 595 m. and that of Ko-usu as
580 m., while the Imperial Geological Survey reports 736 m. and
609 m. respectively. This difference is too great to be regarded as
an error in surveying and must mean that some igneous intrusion
produced the irregular change of elevation. The writer presumes
that the present height of Usu would be found to differ materially
from that recorded by the Imperial Geological Survey.
One year after the eruption Dr. Omori’ observed that there were
local elevations and depressions of the ground in the vicinity of
the mountain and over an area of 150 sq. km. The Military
Survey Department undertook the determination of height at the
request of Dr. Omori and found that Mount Usu was raised, while
the western foot was depressed. In the following year the same
surveyor recorded contrary results; that is, the previously ele-
vated portion had been depressed, while the depressed part was
uplifted.
VI. SUMMARY
1. Earthquakes and roaring before an eruption.—As a rule the
eruption of Usu is preceded by the foreshocks. This, in the opin-
ion of the writer, suggests that the lava reservoir was located nearly
at the same depth in the case of the recent eruptions. The magma
in the reservoir, becoming highly heated, could not retain the
involved gases, and so the maximum strain under the crust was
produced by the continuous heating process. The explosion took
place when the interior and exterior pressures were not counter-
balanced. ‘Thus the pressure of the highly heated magma over-
balanced both the atmospheric and the crustal pressures. The
ground beneath the surface burst, owing to the intense strain, and
produced the loud sound. The speed of the earthquake waves
is greater than that of the sound traveling in the ground and the
air. The minute tremor which normally precedes the sound was
not noticed because there was no seismograph at hand. Hence,
t Bulletin of the Imperial Earthquake Investigation Committee, V, No. 3 (1913).
INTERESTING PHENOMENA ON THE ERUPTION OF USU 285
in a seemingly contradictory fashion, the large tremor of the
earthquake was felt after the sound was heard.
2. Least resistance——From the structural point of view the
greatest number of fractures were observed along the sides of the
great depression, or along the anticlinal top; and especially large
fractures were found close to the edge of the depression. On the
coast line of the Pacific Ocean the presence of several volcanic
_ Fic. 25—‘‘New Mountain” almost completed. Taken October 25, 1910
zones naturally demonstrates the existence of fractures made by
the depression of the Pacific basin. Two recent faults in the
vicinity of Usu were made by the earthquakes, many fractures
usually accompanying the fissures along the aperture. Coming
back to the original Lake Toya, the writer believes that the depres-
sion of the ground produced the lake, which is surrounded by com-
paratively sharp cliffs; as T. Kato stated in his report,’ there must
_ have been some fractures along the margin of the lake through
which Mount Usu erupted. Such fissures, the writer dares to say,
Report of the Imperial Earthquake Investigation Committee, No. 62.
286 Y. OINOUYE
are the weak lines around the foot of Mount Usu. The gases
involved within the magma found an exit through the old fissures,
the lines of the least resistance, which existed under the lava-flow,
and thus the two fissure zones, parallel to the shore line of the lake,
were made. As in the case of a viscous substance which is being
boiled and shows the evolution of gases in certain restricted points
which migrate around over the surface of the liquid, we may assume
that the gases evolving from the magma are generated at different
Kite-Byobu-Yoma
New Mountain
Nishi-Mavu-Yara
Fic. 26.—Topographical comparison of the north side of the Usu. Broken line,
the slope before the elevation. Taken July 29, 1910. Full line, the present relief.
Taken October 25, 1910. Dotted line, the actual difference of altitude.
places without reference in time to each other. The irregular
eruption of the craterlets may be explained on this hypothesis.
The independent activity of the new craterlets to the old, small,
steaming pits on the O-usu and Ko-usu suggests by their lack of
sympathy that they do not rise to the surface through the same
lava vents and that the reservoirs are not connected.
3. Origin of the “‘New Mountain.”—Professor F. Omori’ stated
that the ‘‘New Mountain” is due to the intrusion of lava in the
form of a spine or dome, and Professor D. Sato? believes the intru-
* Bulletin of the Imperial Earthquake Investigation Committee, V, No. 1, p. 1.
2 Bulletin of the Imperial Geological Survey of Japan, XXIII, No. 1 (1913).
INTERESTING PHENOMENA ON THE ERUPTION OF USU_ 287
sion to be a laccolith. Many geologists agree with the theory of
G. K. Gilbert as to the formation of the laccolith. It is a plausible
supposition that the propelling magma would find the line of
least resistance in certain planes, lifting the land above it. In the
early stages of the activity in Mount Usu enormous quantities of
gases were emitted, together with ashes and bombs, while at the
mature stage the north side of the second zone of craters was
sharply elevated in a straight line. Ernest Howe" made experi-
ments on the intrusion
of wax into plaster,
marble, sand, and coal
layers, and demon-
strated how the lacco-
liths are formed. Where
there are fissures from
the inner source to the
surface, there must bea {|S
line of least resistance Fic. 27.—A diagram showing the intrusion of
at this place. Intrusion “plug”: A, B, C, D, mountain slope before the
between the strata eruption; B, E, F, elevated mountain; C, shore
line before the eruption; F, present shore line.
occurs only where there
are no fissures or cracks extending to the surface. It is unreason-
able to believe in the intrusion of the laccolith while the distinct
cracks shown in the two zones of craterlets are in evidence, as we
have seen in the experimental data of Howe. The majority of
nearly 150 laccoliths in the western part of the United States of
America are composed rather of acidic rocks, while rock? as basic
as that of Usu is found only in rare cases. The basic lavas pre-
serve a comparatively high degree of liquidity down to rather low
temperatures, with a quick process of solidification by rapid crys-
tallization, as is well illustrated in blast-furnace slag. Ejection of
many bombs demonstrated that the lava was not seated in the
great depths.
From the facts stated above, the writer is inclined to believe that
the formation of a ‘‘plug,” elongated west-northwest by east-
t Twenty-first Ann. Rept., U.S.G.S., Part III (1901).
2 See the analysis of bomb.
Old Lava Flow
288 Y. OINOUYE
southeast and elliptical in plan, took place in the midst of the
activity. Of course, we must not forget that the elevation is not
entirely due to the intrusion of the plug, but that there was co-
operation of the faulting such as is so remarkably shown on the
top of Kompirayama. The elevation of the lake shore on the
south side of Lake Toya may be accounted for by the tilting of the
crust owing to the intrusion of the plug rather than, as previously
supposed, to the formation of a laccolith (Fig. 27). From dynamic
considerations it is evident that if the force of the plug intrusion be
applied to one end of a resistant section of the earth’s crust the
whole block will be hfted and tilted.
How could such a very steep slope (40°-70°) on the south side
and 22°-30° on the north side be made on the surface by the intru-
sion of a laccolith? If we suppose that there is a very sharp, steep
dome in the great depth overlaid by heavy layers above, its inclina-
tion becomes gradually gentle toward the surface, unless the crust
be in the liquid or semiliquid condition.
4. Undulation of ground near Usu.—From the damage done, and
from faults, fissures, and mud cones which were found exclusively
on the same side, we may prove that the structure of the western
region was originally weak, so that the shaking was intense, while
in other parts the effects were comparatively small. Frequent
explosion also weakened the already feeble lines.
We may then conclude, from evidence gathered in the field, that
there must have been intruded irregular bodies of lava that pro-
duced the undulation of the ground near Usu recently observed in
the eruption.
INTRAFORMATIONAL PEBBLES IN THE RICHMOND
GROUP, AT WINCHESTER, OHIO
AUGUST F. FOERSTE
Dayton, Ohio
Ripple-marks, measuring two feet or more from crest to crest,
occur at numerous horizons in the Ordovician rocks of Ohio,
Indiana, and Kentucky, but are abundant especially in the middle
parts of the Richmond group, where they characterize the upper
part of the Waynesville formation and the lower part of the imme-
diately overlying Liberty formation. Among the occurrences of
ripple-marks discussed by Prosser, in his recently published paper
on the ‘“ Ripple-Marks in Ohio Limestones,’”* the following belong
to the lower part of the Liberty formation: Elk Run, a little over
a mile east of Winchester, Ohio (Figs. 1, 2); Cherry Fork, at
Harshaville, 6 miles east-southeast of Winchester; and Treber
Run, 5 miles southeast of Harshaville. The ripple-marks described
by Joseph Moore and Allen D. Hole from a small western tributary
of the Whitewater River, 5 miles southwest of Richmond, in
Indiana, and those described by W. P. Shannon from the bed of
Salt Creek, 3 miles west of Oldenburg, 38 miles southwest of Rich-
mond, also occur in the lower part of the Liberty formation. At
the Ridenour Mill, 53 miles northwest of Oxford, Ohio, the ripple-
marks described by Nelson W. Perry” occur both in the lower part
of the Liberty and in the upper part of the underlying Waynesville
formation. In fact, over a large part of Ohio and Indiana, ripple-
marks are fully as abundant in the upper part of the Waynesville
as in the lower part of the Liberty, and the list of localities might
be multiplied almost indefinitely. Ripple-marks occur also near
the top of the Brassfield formation at numerous localities in south-
ern Ohio (Figs. 3, 4) and northern Kentucky, east of the Cincin-
nati axis.
t Jour. Geol., XXIV, No. § (1916), pp. 450-57.
2 Am. Geol., IV (1889), 326-36.
289
bo
go AUGUST F. FOERSTE
Fic. 1.—View of ripple-marked limestone in bed of Elk Run, looking northward
toward abutment of road bridge. The appearance of strong contrast in the slope on
opposite sides of the ripples is due to the dip of the rock toward the right causing the
water to rise higher on the left side of the crests. In reality, the difference in slope on
the two sides is small.
Fic. 2.—View at the same locality, looking northeast across the stream. See
also Fig. 2, on p. 460 in Journal of Geology, XXIV, No. 5 (1916).
INTRAFORMATIONAL PEBBLES IN RICHMOND GROUP 2091
Fic. 3.—Ripple-marks on Brassfield limestone in Beasley Fork, a mile and a
quarter south of West Union, Ohio. Note the small difference in slope on the two
sides of the ripples.
Fic. 4.—Ripple-marks at the same locality and horizon on Beasley Fork
292 AUGUST F. FOERSTE
Occasionally pebbles occur in the ripple-marked layers of lime-
stone. ‘These pebbles usually are few in number and rarely are
sufficiently abundant even to suggest the term conglomerate.
They are more abundant at two horizons in the lower part of the
Liberty formation at the locality on Elk Run, east of Winchester,
Ohio, described by Prosser, than at any other localities known at
present in the Ordovician of Ohio, Indiana, and Kentucky, and there-
fore this locality has been chosen to present some of the features
characteristic of these pebbles. The pebbles are of two types.
t. In one type the rock is very fine-grained, as though originat-
ing from a calcareous mud, and is frequently marked by worm-
burrows. ‘There are also peculiar gouged-out markings, 3 or 4 mm.
wide, 60-100 mm. long, often 20 mm. deep at the center, curving
downward from the ends toward the center as though carved out
by some narrow gouge. Markings of this kind frequently connect
the two oblong or nearly circular terminals forming the peculiar
dumb-bell fossil called Arthraria. There are also gouged-out
markings only an inch in length (Fig. 6, pebble C). The surface
of these pebbles often is very irregularly rounded, as though the
rock had been soft at the time of formation of the pebbles. At
one horizon these pebbles frequently support small colonies of the
incrusting coral Protarea richmondensis (Fig. 5), 40-70 mm. in
width. In fact, twenty pebbles supporting Protarea richmondensis
were exposed along a narrow outcrop, a foot wide and scarcely
50 feet long, at the time of my last visit. Young specimens of
Streptelasma, presumably Streptelasma rusticum, to-15 mm. in
length, occasionally occur, attached by their sides to the pebbles.
Three pebbles supporting young specimens of Streptelasma occurred
in the s5o0-foot length mentioned above. Incrusting growths of
Dermatostroma corrugata and of various thinly incrusting bryozoans
also occur occasionally. Since the incrusting growths follow the
irregular curvature of the pebbles, it is evident that the latter is
not due to subsequent erosion. Jn one specimen, thinly incrusting
bryozoans and growths of Siomatopora occur on the lower side of
the pebble, while thicker growths of Protarea richmondensis occur
on the upper surface, showing that the pebble had been turned
over at least once.
INTRAFORMATIONAL PEBBLES IN RICHMOND GROUP 293
The fine-grained mud of which these pebbles are composed
incloses but few fossils. In three pebbles Lophospira bowdeni
was found, and one pebble contained a ventral valve of Dinorthis
subquadrata, a characteristic fossil of the Liberty formation,
unknown in the underlying Waynesville strata. In other words,
there is no reason for believing that the rock of which the pebbles
are composed is older than that of the formation in which the peb-
bles now occur. Im fact, at several localities along the creek, the
rock immediately underlying the pebble-bearing layer is sufficiently
similar to the rock forming the pebbles to have given origin to the
latter.
These fine-grained pebbles usually do not exceed 6 inches in
length, 3 inches in width, and an inch in thickness, but specimens
12 inches long, 7 inches wide, and 2 inches thick are known, and
one pebble 18 inches long, 11 inches wide, and 3 inches thick was
observed.
2. The second type of pebbles usually consists of a fine-grained
blue limestone, in which the grain is distinctly less fine than in the
first type. The granular structure usually may be recognized with-
out the assistance of a magnifier. Worm-burrows usually are
absent, and no incrusting bryozoans or specimens of Protarea have
been observed. The color of the rock is bluish gray, similar to that
of the inclosing rock, and the outlines of the pebbles may be dis-
tinguished from the latter chiefly by the finer grain of the rock
forming the pebbles, and usually also by differences in the strati-
fication planes running through the rock. Such fossils as occur
in these pebbles suggest the Liberty age of the rock from which the
latter were derived, and the source of this rock could have been one
of the underlying layers within this formation.
The pebbles of this second type usually are relatively thin and
flat. The upper and lower surfaces usually are parallel, the lateral
margins often being vertical or rounding only moderately into the
upper and lower surfaces. In vertical cross-sections, therefore,
the pebbles appear angular at the margins. Angularity fre-
quently characterizes also the lateral outlines, as observed from
above. In other words, the pebbles frequently appear broken off
from thin layers of limestone, without much rounding. Some of
204 AUGUST F. FOERSTE:
the pebbles are only 4 inches in length, and only a few exceed
12 inches in length, 7 inches in width, and 1 inch in thickness,
but occasionally specimens much larger than this are seen. One
Fic. 5.—Pebble, of natural size, supporting two colonies of Protarea rich-
mondensis. From layer D in the Elk Run section, east of Winchester, Ohio.
295
INTRAFORMATIONAL PEBBLES IN RICHMOND GROUP
‘VY 3® pozeo0] St
‘oly ‘a1sayou
I
°
M
JO 4sva ‘UOT}O—aS uny MA 94} UL Ci Joke] Wory = ‘uorpisod payszsaur ue ut yng
Aq paquasaidas auo ay} asay} Jo ‘satqqod yeszaaos sunsoddns ‘azis peinyeu patyj-au0 jo ‘quis auojsaIJ—"9 ‘onT
206 AUGUST F. FOERSTE
pebble 38 inches long, 28 inches wide, and 2.5 inches thick was
found. :
The relative position of the pebble-bearing layers among the
ripple-marked limestones along Elk Run may be seen from the
following section. Ripple-marked layers occur here at various
levels in a section at least 30 feet thick. Owing to the fact that the
dip of the rocks is in the same direction as the flow of the stream,
several layers disappear below the level of the stream at one point
and reappear farther down the stream. ‘This makes the unraveling
of the section more or less difficult in places, but the following sec-
tion, described in descending order, is as nearly correct as may be
determined from the present condition of the exposures, which is
unusually favorable for this locality.
SECTION ALONG ELK RUN, 1.5 MILES EAST OF WINCHESTER,
OHIO
Ft. In.
Layer A. Exposed about 600 feet south of railroad bridge; crest
oi ripples running N. 47° W. Pebbles few.
HD ol 5 mec Wa ree ge Aan ae, MR pacer ciara aa ere MUR nu Ken a) Ss 2 5
Layer B. Crest of ripples running N. 40° W.
JET RiCeh age Ban AAU RC ee ane ep NE Ny UM ReMi NS ote Sabo Br 3 9
Layer C. Crest of ripples running N. and S.
| Gs es geet UB LU Mara el DIC aa AA La SLE ORR POURS UAL CN 9
Layer D. Crest of ripples running N. 40° W. This layer is not ripple-
marked southeast of the home of Charles Bailey, but here
numerous fine-grained pebbles, many supporting growths of
Protarea richmondensis, occur (Figs. 5, 6).
Tmtervall 3 esis eee Gei Na CD rach oat ANAT Mica ical a SUES a eae mate eer I
Layer E. Crest of ripples running from N. 30° W. to N. 25° W.
Peterviadi ae i eae irene Ute ia TIN U3) Rs NM nna ese 9
Layer F. Crest of ripples running N. 35° W.
Erste rv ey hac ral eet aay RCE ses Ral iekl er aw uae ay aig I 6
Layer G. Crest of ripples varying from N. 3° W. to N. 15° W. at the
small fall northeast of the home of Charles Bailey. This
layer is exposed also immediately north of the road bridge,
and between the road bridge and the railroad bridge. It
is characterized by angular, flat pebbles, few in number,
but sometimes of considerable length and width, consider-
ing the small thickness.
UB ye) uigcll LAPP pee oan At ULSTER a a RE RSCU I ean) URC ul eho 6
INTRAFORMATIONAL PEBBLES IN RICHMOND GROUP 2097
Layer H. Crest of ripples running N. 15° W.
lintenviall ener a eco nasi ougers oe uii a TU oho e al ala ule 5
Layer I. Crest of ripples running N. and S., faintly defined. Near
home of E. E. Jamison.
LEOY ETN oy cb GloN aan le eve: CAMO S2 pay RS Ro 5
Layer J. Poorly defined ripple-marks, at next house on west side of
creek.
ace tayallee eye ee ena teeta Ma RUE UMC tei ela te ited I
Layer K. Crest of ripples running N. 70° E.
J Ura ies au A even een AST SS aes Neat LO OG ee I
Layer L. Crest of ripples running N. 50° E.
The fine-grained pebbles first described, apparently consisting
of a lime mud, supporting incrusting growths of Protarea, Dermato-
stroma, and various species of bryozoans, are specially characteristic
of layer D. This layer is exposed at several localities along the
creek, but the pebbles are common only southeast of the home of
Charles Bailey, on the eastern side of the creek. Here the pebbles
rest upon the top of the layer or are more or less imbedded in its
upper part. This pebble-bearing layer is exposed also farther up
stream, about 500 feet north of the road bridge. Here the pebbles
vary from 2 to 4 inches in length, and from a quarter to half an
inch in thickness. Farther up stream, immediately south of the
road bridge, this layer is strongly ripple-marked, the crest of the
ripples running N. 40° W. One pebble was noticed here 14 inches
long, 7 inches wide, and half an inch thick. Farther north, where
the pebbles are abundant, ripple-marks are absent.
The less fine-grained and more angular blue limestone pebbles,
described last, occur in layer G. This layer is exposed between the
railroad bridge and the road bridge, a short distance northward.
The crests of the ripple-marks vary here (Figs. 1, 2) between N. 3° W.
and N. 15° W. in direction, and on the average are about 30 inches
apart. The following pebbles were noticed here, imbedded within
the upper part of the ripple-marked layer, only the upper surface
being exposed. In each case the length, width, and thickness of the
pebble are given. One pebble, 4X3Xo0.5 inch, lay in a trough
and sloped gently toward the west. Another, 12X7Xo.5 inch,
lay in a trough in a horizontal position. ‘Two pebbles, 4x 4Xo. 25
inch, lay in a trough and sloped toward the east. Two pebbles,
298 AUGUST F. FOERSTE
4 inches in transverse diameter, lay in a trough in a horizontal
position. One pebble, 8X 5 Xo. 5 inch, was imbedded on the eastern
side of one of the crests, but in a horizontal position. ‘Two pebbles,
one on the eastern and one on the western side of the same crest,
were in a horizontal position. The steeper side of the ripple-marks
lies on the western side of the crests. The same layer is exposed
immediately north of the road bridge, with layer D about 3 feet
3 inches farther up.
Layer G is exposed again at the small waterfall northeast of the
home of Charles Bailey. The crests of the ripple-marks here run
N. 3° W., the crests are about 28 inches apart, and the steeper slope
is on the western side. Here the following pebbles were noticed,
the dimensions being given in inches: one pebble, 1oX 7X1 inch,
in a horizontal position, imbedded along the crest of one of the
ripples; a pebble, 12X71 inch, in a horizontal position, buried
under the western half of a crest; a pebble, 38X28X 2.5 inches, in
a horizontal position, imbedded so that its upper surface is on the
same level as that of the surrounding rock. The moderately rounded
margins are slightly overlapped by the surrounding rock; and the
ripple-marks characterizing the latter are clearly defined as far as
the margin of the pebble, but are absent, of course, on the surface
of the latter. f
Layer G is exposed also farther down the stream, northward.
Here the crests of the ripples run N. 15° W., both sides of the ripple-
marks sloping equally. One pebble, 6X 4X1 inch, was imbedded
up to its upper surface within the ripple-marked layer, and an
incrusting bryozoan overlapped one margin of the pebble and the
adjacent part of the surrounding rock, showing that enough time
elapsed before the deposition of the overlying clay bed to admit
of the growth of this bryozoan, the thickness of the latter being
about 3 or 4mm.
The pebbles in layer A were few in number. One pebble,
6X4Xo.5 inch, consisted of fine-grained rock, resembling the worm-
burrowed layer beneath the ripple-marked limestone. Farther
north, nearer the railroad bridge, several additional pebbles,
consisting of the same kind of rock, were found. The crests of
the ripples run N. 47° W., they are 20-30 inches apart, the inter-
INTRAFORMATIONAL PEBBLES IN RICHMOND GROUP 299
vening troughs are about 2.5 inches deep, and the steeper slope
is on the western side.
All of the strata included within the 30-foot section here
described belong to the Liberty formation. The lowest layer, L,
contains the characteristic fossil Dinorthis subquadrata, and Plec-
tambonites sericea is so abundant here that it suggests a horizon
not far above the base of this formation. The abundance of typi-
cal Strophomena planumbona throughout the section suggests the
lower half of the Liberty. Strophomena vetusta, associated with
Dinorthis subquadrata and RKhynchotrema capax, is comparatively
rare until the layers immediately overlying layer A are reached,
but the general aspect of the rock here still is that of the Liberty
formation.
Judging from exposures on Graces Run, a little over a mile
west of Harshaville, the highest strata exposing ripple-marks occur
at least as far up as within 88 feet of the base of the Brassfield
limestone. Pebbles up to 6X4Xo.5 inch in dimension occur at
a small fall half-way between this point and the mouth of Martins
Run, half a mile southeastward. Several pebbles occur also in
the wave-marked layers in the bed of Cherry Fork, immediately
west of Harshaville.
The highest ripple-marked horizon along Elk Run, east of Win-
chester, appears to be about 80 feet below an exposure of Brassfield
limestone seen along the railroad, west of the creek.
Pebbles up to 7X 4Xo.5 inch in dimension occur also on Treber
Run, about a quarter of a mile west of the mouth of the stream, a
short distance west of the first crossing of the road following the
stream. Here the pebbles occur in large, loose slabs of limestone
containing a Liberty fauna, and evidently not transported far.
_ The pebbles consist of small-grained blue limestone, similar to those
occurring in layer G in the Elk Run section.
Perhaps the chief reason why the presence of pebbles in these
ripple-marked strata has received so little attention is because they
are so readily overlooked. By far the larger number of pebbles
are horizontal in position, their stratification planes coinciding in
direction with those of the inclosing rock. Especially is this true
of the larger pebbles, while the smaller pebbles occasionally occur
300 AUGUST F. FOERSTE
at distinct angles with the inclosing rock. Moreover, the upper
surface of the larger pebbles rarely projects distinctly above the
surface of the inclosing rock, but more commonly is about at the
same level as the latter. On careless examination, the pebbles
appear merely as adhering remnants of the next overlying layer of
rock. They are distinguished chiefly by the finer grain of the peb-
bles, frequently accompanied by a difference in color and by a
difference in the character and location of the stratification planes.
There is no evidence that the larger pebbles, a foot or more in
diameter, ever were turned over so as to present the lower instead
of the upper surface of the original rock stratum. Perhaps this
statement could be made with equal accuracy of any pebble 6 inches
in width. The largest pebbles, so far found, which give evidence
of having been overturned before being imbedded have a width of
almost 3 inches, although the length may equal 6 inches. In one
of these pebbles, obtained from layer D in Elk Run, a thin growth
of Ceramoporella ohioensis on one edge of the pebble overlaps both
the upper and the lower surface of the latter by fully an inch, and
additional growths of the same species occur on the lower surface,
the entire width of the pebble being 2$ inches. The incrusting
specimens of Protarea frequently occur on the upper surface of the
pebbles, often several colonies on the same pebble, and these colonies
frequently overlap the lateral edges, but never occur on the lower
surface of the pebbles.
The size and the angularity of the larger pebbles suggest that
they have not been transported very far. The absence of over-
turning of these larger pebbles also suggests only a short distance of
transportation. The very irregular surface features of the very
fine-grained pebbles, among which evidences of overturning are
more frequent, suggest washed lumps of partly indurated calcareous
mud rather than strongly eroded and frequently overturned rock.
Although rock similar to that forming the pebbles frequently
occurs immediately below the layer in the upper surface of which
the pebbles are imbedded, these ripple-marked layers frequently
are continuous over such large areas, as determined from exposures
along the lateral branches of streams, as to make the origin of the
pebbles from underlying strata more or less doubtful. Especially
INTRAFORMATIONAL PEBBLES IN RICHMOND GROUP 301
is this true in the case of the larger pebbles, which apparently
have been transported only short distances.
Moreover, there is no evidence of strong unconformities any-
where in the Richmond series of rocks. At no point has a layer
of rock been found to overlap the lateral margins of any of the
layers for even a vertical distance of 2 feet. Hence an origin from
anything like a cliff or coast or beach seems questionable. At
least there is no evidence of the presence of any cliff, coast, or
beach sufficiently close to the area in question to have furnished
the material for the pebbles.
There was a tendency formerly to regard the presence of ripple-
marks as evidence of shallow-water conditions and as suggesting
the proximity of shore lines. This found expression in a paper
by Joseph F. James on the “Evidences of Beaches in the Cincin-
nati Group.”! Here the ripple-marks in the upper part of the
Cynthiana formation, at Ludlow and West Covington, Kentucky,
opposite Cincinnati, and another set of ripple-marks about 300
feet above low-water mark in the Ohio River, presumably in the
upper part of the Eden formation, were interpreted as evidences
of the proximity of beaches. As further evidence of shallow-water
conditions during the deposition of various parts of the Cincinnati
group, the presence of raindrop impressions near the top of the
Cincinnati group was cited, but the exact location of the rock
bearing these raindrop impressions is not given.
The impression that at least a part of the rocks of the Cin-
cinnati group were deposited in very shallow waters finds expres-
sion also in a paper by Nelson W. Perry on “The Cincinnati
Rocks; What Has Been Their Physical History?’ Here rain-
drop impressions are cited from the vicinity of Smiley’s Dam, 33
miles southeast of Oxford, but 5 miles distant when approached by
the road. The dam is located on Fourmile Creek. The lowest
strata exposed belong to the Mount Auburn division of the Mays-
ville formation, and a mile westward the Dinorthis carleyi horizon
near the middle of the Arnheim bed is exposed at an elevation about
50 feet higher. Careful search by the present writer failed to locate
t Science, V (1885), 231.
2 Am. Geol., IV, No. 6 (1889), pp. 326-36.
302 AUGUST F. FOERSTE
the presence of raindrop impressions or of any other evidences
suggesting deposition under shallow-water conditions.
Perry next calls attention to the exposures at the Ridenour Mill,
on Little Fourmile Creek, about 7 miles north-northwest of Oxford,
where numerous layers in the upper part of the Waynesville and
lower part of the Liberty divisions of the Richmond group are
ripple-marked.
Next, Perry cites the presence of mud-cracks from a locality
near Moores Hill in Dearborn County, in Indiana, presumably
from the Waynesville division of the Richmond group. And,
finally, he alludes to the well-known ripple-marks in the upper part
of the Cynthiana formation, at Ludlow, Kentucky.
Now, whatever may be the opinion concerning the value of
ripple-marks as evidence of shallow-water conditions, there can
be no difference of opinion as to the evidence presented by rain-
drop impressions and mud-cracks. However, the presence of
raindrop impressions and of mud-cracks must be fully proved.
This the present writer has been unable to do.
Specimens formerly interpreted by him as exhibiting rain-
drop impressions he now regards as ripple-marked, irregular
ripples of short amplitude crossing at various angles, leaving
intermediate more or less circular hollows. If anyone has clear
evidence of the presence of raindrop impressions in Cincinnatian
strata, this evidence should be published, accompanied by clear
illustrations.
As regards the presence of mud-cracks, the present writer has
seen many occurrences of structures suggesting mud-cracks, but
has come to regard their origin from exposure of mud-flats to aerial
conditions as extremely doubtful.
When mud exposed to the drying effects of the open air cracks,
it not only tends to pull apart at the cracks, but the upper, more
rapidly drying part tends to pull away from the part beneath.
Frequently the cracked surface becomes sufficiently hardened to
remain more or less intact when the next tide proceeds to cover it.
This causes the subsequently deposited material to settle in part
in the cracks, and frequently the part filling the cracks is sufficiently
different to be readily distinguished from the original mud deposit.
INTRAFORMATIONAL PEBBLES IN RICHMOND GROUP 303
In the case of sea deposits, the material filling mud-cracks might
include coarser-grained deposits or organic material, either entire
or fragmental. Now, it is the very frequency of the supposed mud-
cracks with the absence of the concurrent phenomena here indicated
which. throws doubt on their interpretation. In such ‘mud-
cracks’? as have been observed hitherto, the material filling the
cracks is essentially the same material as that forming the lateral
walls of the crack, and no fossil or fragmental material has ever
been found in a position suggesting that it had been dropped into
the crack, or had been washed into it.
On the contrary, in many cases it has appeared possible that the
cracking could have occurred subsequent to the deposition of the
overlying strata, in fact long subsequent to the latter, and may not
be due to the drying effects of the air along a seashore, but to shrink-
age of strata deposited in much deeper waters. Mud deposits
in quiet waters have been known to crack without exposure to the
air, although the observed cracks have always been of too small
magnitude to suggest mud-cracks. Shrinkage, however, may have
occurred also long subsequent to the deposition of the overlying
strata, during a period of elevation of the entire mass of marine
deposits. The gradual filling of the cracks might have been accom-
plished by slowly circulating waters while the shrinking material
still was comparatively soft. While the method of filling of these
cracks may vary in different cases, the possibility of their origin
from shrinkage long after the deposition of the strata in which they
occur should be considered. If anyone has any evidence of the
presence of mud-cracks in Cincinnatian rocks which unquestion-
ably are due to elevation of mud flats above water-level before the
deposition of the immediately overlying strata, this evidence should
be published in detail.
Until the presence of raindrop impressions and of mud-cracks
due to exposure of mud-flats to the open air before the deposition
of the overlying strata has been proved unequivocally, it is not so
certain that ripple-marks indicate shallow-water conditions. They
may have been formed a considerable distance below sea-level, at
least sufficiently far not to necessitate the immediate presence of a
shore line.
304 AUGUST F. FOERSTE
Formerly, pebbles were regarded as unequivocal evidences of
the proximity of a shore line, but even these may be formed below
water-level. This is true especially of the pebbles found in the
Ordovician strata of Ohio, Indiana, and Kentucky, since these
usually occur only in ripple-marked layers, or in lateral extensions
of these layers. The same causes which gave rise to the ripples
may have given rise to the pebbles.
These causes apparently include a more or less rapid flow of
water. The ripple-marked layers usually consist of more or less
fragmental detrital or organic material, frequently in much greater
quantity and of coarser grain than in the immediately overlying
and underlying layers, as though freed from the accompanying
calcareous and argillaceous muds by repeated rewashings of the
materials constituting the ripple-marked layers. These muds
either were washed to more distant areas or were, in part, held in
suspension in the overlying waters for a short time. The ripple-
marked layers frequently show evidences of cross-bedding, espe-
cially immediately beneath the crests of the ripples, thus also
suggesting current action. The larger pebbles, a foot or more in
diameter and only an inch or two in thickness, may easily have
been formed by currents dissecting a more or less fine-grained
stratum, and leaving remnants of the latter more or less imbedded
in the current-washed material farther on. In limestone layers
not exceeding four inches in thickness, even directly beneath the
crests of the ripples, the larger pebbles scarcely could incline very
much. The source of the pebbles readily could have been some
formerly existent layer located less than a foot above the present
level of the pebble. The finer muds of the intervening section
could have been washed away, and the coarser material retained
to form the major part of the ripple-marked layer, in the upper
surface of which the limestone pebbles are imbedded.
Especial attention should be called to the fact that, even where
the ripple-marked layers are most abundant, many of the inter-
mediate limestone layers may show no trace of ripples. Hence, the
frequent absence of ripple-marks needs explanation fully as much
as their locally more or less frequent presence.
INTRAFORMATIONAL PEBBLES IN RICHMOND GROUP 305
One of the causes giving rise to widespread current-action may
have been violent and widespread storms. Considering the fact
that the Cincinnatian strata were deposited in epicontinental
seas, storms easily, at times, might have blown vast quantities of
water over those parts usually covered only by shallow waters.
Such storms occasionally are experienced on the Gulf coast, and
along the coast of the southern Atlantic states. The ebb flow
of these accumulated waters, after the storm, might easily give rise
to widespread ripple-marking of the last deposited strata, even at
a considerable distance from actual shore lines. Such an origin
of currents would predicate wide areas of gradually shallowing
seas over which the surface waters blown before the wind would
tend to accumulate. Currents due to such causes might be
expected more readily in the comparatively shallow waters of
epicontinental seas than along the more abrupt shores of the deeper
oceanic basins.
The presence of numerous well-preserved colonies of Protarea
and of delicate growths of Stomatopora and other bryozoans on
the upper surface of the pebbles at the Elk Run locality, east of
Winchester, Ohio, is indicative of submerged conditions at least
immediately after the formation of the pebbles. In a similar man-
ner, the long crinoid columns found on the surface of the ripple-
marked layers of lower Trenton limestone, at Hull, in Canada,*
indicate that fairly deep waters were present immediately after the
formation of the ripple-marks, and may have been present even
during their formation.
Ripple-marks comparable in dimension with those character-
istic of the Ordovician limestones of Ohio, Indiana, and Kentucky
are not unknown along our present shores. They occur where
waters accumulating in extensive salt marshes or in estuaries during
high tide find a ready outflow to the sea during ebb conditions.
In these cases the steeper slopes of the ripple-marks are directed
away from the shallower waters.
It is remarkable, however, in the case of the ripple-marks on
the Ordovician limestones of the areas here under discussion, how
t Kindle, Jour. Geol., XXII (1914), 712.
206 AUGUST F. FOERSTE
slightly the slopes on the two sides of the crests of the latter differ
in most cases. Frequently it is difficult to determine a difference
in slope at all, and rarely is this difference strongly defined, as in the
case of the strong ripples found after ebb tide along our present
coasts.
Eventually it may be possible to accumulate sufficient evidence
to determine with considerable certainty the conditions under
which many of the strata of ancient days were deposited. To
many the evidence appears to be already at hand. To others a
revision of the evidence may appear necessary. To the writer it
appears desirable that those who have indubitable evidence of
land conditions during the deposition of Ordovician strata in the
states of the Ohio Valley should publish the same. This is true
especially in case of raindrop impressions and of mud-cracks,
which are favorite evidences locally of shore and land conditions
during Ordovician times.
REVIEWS
Upper White River District, Yukon. By D. D. Carrnes. Geol.
Survey Canada, Memoir 50, 1915. Pp. 1o1, figs. 2, pls. 17,
maps 3.
This report covers an area of about 800 square miles lying along the
Alaska-Yukon International Boundary from latitude 61° 40’ to 62° 30’.
It is considered to be a promising area for mineral deposits of economic
value.
The oldest rocks exposed are mica schists referred to the Yukon
group of pre-Cambrian age. Upon these rest 1,500 feet of Carbonifer-
ous limestones and clastics followed by 1,000 feet of Mesozoic shales
and sandstones. At a few points Tertiary beds were observed. These
beds are in part flat-lying, and in bart have been highly dynamically
metamorphosed.
The writer believes that the Nutzotin Mountains are due to differ-
ential erosion rather than to faulting. They remained as a region of
considerable relief at the time of the peneplanation of the Yukon plateau
region and were further uplifted between the late Miocene and the early
Pleistocene. A different explanation from that suggested by geologists
of the United States Geological Survey is advanced to account for
drainage changes along White River.
: : W. B. W.
Wyoming and McDowell Counties. By R. V. HENNEN. West
: Virginia Geol. Survey, 1915. Pp. 783, pls. 31, figs. 28, maps 2.
McDowell County, situated on the southern border of the state, has
led all the counties in the state in coal production since 1905. Approxi-
mately 15,000,000 tons were produced in 1915, and at this rate its avail-
able coal will last about two hundred and fifty years. Wyoming County
coal fields have not been developed until recently, but its coal reserves
equal those of McDowell County.
The Pottsville series has a remarkable development here. It
increases from a thickness of 250 feet at the northern edge of the state
to'a maximum of 3,850 feet in these counties. It has been differentiated
into three groups and two score formations.
397
308 REVIEWS
A new feature in this report is a series of 25 maps of these counties
showing the minable coal areas of as many different coal horizons.
Under separate cover are topographic and geologic and structure contour
maps.
W. B. W.
Ou and Gas Fields of Ontario and Quebec. By WYATT MALCOLM.
Geol. Survey Canada, Memoir 81, 1915. Pp. 248.
This memoir has been prepared chiefly for those interested in the
commercial development of oil and gas. It treats of the lithology,
stratigraphic relations, and areal distribution of the geologic formations
from the Potsdam to the Chemung. The predominant structural fea-
ture is a gentle dipping of the strata to the southwest away from the
pre-Cambrian axis. The northeastern extension of the Cincinnati
anticline reaches into Ontario.
The productive horizons are not limited to one formation. Gas is
found in the Medina, oil and gas in the Guelph and Salina, and oil in
the Onondaga. The production of gas has increased steadily, but the
oil output reached a maximum in 1907, and since then has fallen greatly.
Analyses of gas from different fields show a surprising uniformity of
composition. The writer of the report believes this to be incompatible
with a local and separate origin of the gas for each field.
W. B. W.
Arisaig-Antigonish District, Nova Scotia. By M. Y. WItLIAms.
Geol. Survey Canada, Memoir 60, t914._ Pp. 173, maps 2.
The chief interest in this memoir lies in its contribution to stratig-
raphy. Careful attention had been given already to the region, for
in it lies the key to the stratigraphy of a considerable area. The purpose
of this investigation was to work out in still greater detail the sedi-
mentary record and the ages and relations of the igneous rocks.
Of the Paleozoic systems, the Cambrian and Permian are missing.
Where possible, correlations are made with the type sections of Europe.
Separate chapters are reserved for structural and historical geology.
Igneous geology is given the same careful attention as the sediments.
The igneous rocks are limited largely to acid and basic intrusives in the
Ordovician, and to intrusive diabase sheets in the Mississippian.
W. B. W.
i.
OF THE
By THOMAS C. CHAMBERLIN
Head of the Department of Geology, The University of Chicago
2442 4S
xii-+272 pages, 8vo, cloth $1.50 net, postage extra (weight 1 lb. 6 oz.)
FROM THE PREFACE .
“Tn telling the story of this search for the mode by which the earth came into being, we
have let the incidents that led the inquiry on from one stage to another fall in with the steps of
the inquiry itself. It is in keeping with the purposes of this series of booklets that the motives
which set researches going should have their place with the quests that arose from them. . . . .
The final story of the birth of the earth will come only after a time, when the vestiges of creation
have been more keenly discerned and more faithfully rendered than is possible now.”
- CONTENTS
% INTRODUCTION V. THE ForBippEN FIELp
Ki CHAPTER VI. Dynamic ENCOUNTER BY CLOSE APPROACH
a I. THE GASEOUS THEORY OF EARTH-GENESIS IN VII. THE EVOLUTION OF THE SOLAR NEBULA INTO
c THE LIGHT OF THE KINETIC THEORY OF GASES THE PLANETARY SYSTEM
: II. VESTIGES OF COSMOGONIC STATES AND THEIR VIII. THE JUVENILE SHAPING OF THE EARTH
SIGNIFICANCE TX. INNER REORGANIZATION OF THE JUVENILE
III. THe DeEctsivE TESTIMONY OF CERTAIN VES- EartH
TIGES OF THE SOLAR SYSTEM X. HIGHER ORGANIZATION IN THE GREAT CoN-
IV. Furie Errorts Tact Horizons
One of ‘‘The University of Chicago Science Series’’
THE UNIVERSITY OF CHICAGO PRESS
CHICAGO, ILLINOIS
Why We Send Our i
DAUS IMPROVED TIP TOP DUPLICATOR |
ON TEN DAYS’ FREE TRIAL
FIRST—It proves OUR confidence in the machine. i
SECOND—By personal use YOU can positively tell, before buy-
ing, whether it meets your approval or not. y
Each machine contains a continuous roll of our new “‘Dausco’’ |
Oiled Parchment Back duplicating surface which can be used }
over and over again. Five different colors can be duplicated at ff
the same time. Cleanliness and simplicity of operation and legi- |
bility of copies unequaled by any other duplicating process.
100 Copies from Pen-written and 50 copies PRICE |
= from Typewritten Original. 7 50 i
Take Advantage of Our Trial Offer Complete Duplicator, cap size (Prints 82 X13 in.) e ;
H. DAUS DUPLICATOR COMPANY, Daus Building, 111 John Street, New York |
| FELIX
“i
| Hundreds of Good Books at Bargain Prices
A catalogue containing brief descriptions of each title, and showing
. the original price, as well as our price, will be sent to you without
‘| charge if you will but ask for it.
man | THE BAKER & TAYLOR CO.
Wholesale Dealers in the Books of All Publishers
354 FOURTH AVENUE, NEW YORK AT TWENTY-SIXTH STREET
| FINE INKS 48> ADHESIVES
For those who KNOW
Drawing Inks
Eternal Writing Ink
Engrossing Ink
@ ® 9 Taurine Mucilage
H I g g ins Photo Mounter Paste
Drawing Board Paste
Liquid Paste
Office Paste
Vegetable Glue, Etc.
Are the Finest and Best Inks and Adhesives
Emancipate yourself from the use of corrosive and
ill-smelling inks and adhesives and adopt the
Higgins Inks and Adhesives. They willbea
revelation to you, they are so sweet, clean, well
put up, and withal so efficient.
At Dealers Generally
Just the Typewriter
for You
SOLD ON JUST THE
TERMS THAT SUIT YOU
The Remington Junior
The Junior is smaller and lighter than the Stand-
ard Remington models—weighs only 17 pounds.
It is simpler. You can quickly learn to operate ~
it. No lessons needed.
It has all the Remington essentials, standard key-
board, standard type, and writes letters of stand-
ard size—the kind with the hundred-dollar look. |
Cut out and post today
REMINGTON TYPEWRITER COMPANY
(Incorporated)
327 Broadway, New York
Send me a Remington Junior Typewriter, price $50, on free ex-
amination. It is understood that I may return the machine, if
4
rn
I choose, within ten days. IfI decide to purchase it, I agreeto —
rm
ey
ra
,|
|
pay for itin 10 monthly payments of $5 each.
CHAS. M. HIGGINS & CO., Mfrs.
Branches: Chicago, London
271 Ninth Street Brooklyn, N.Y.
The eye is a short-cut
to the brain. The
Rausch lomb
BALOPTICON
THE PERFECT STEREOPTICON
a short-cut to the eye.
Many progressive educational institutions are taking advantage of these facts and employing
the sharp, brilliant images of the Balopticon as an aid to teaching.
The latest improvement in the Balopticon is the adaptation of the gas-filled Mazda lamp which, though using
much less current than the ‘old-style A.C. arc, furnishes more evenly distributed light and greatly simplifies
operation.
The New Combined Model projects either slides or opaque objects (maps, photos, specimens, etc.), interchange
eee both forms being instant. This model $120. Models for slides only, $25 up. For opaque objects
up.
Special Terms to Educational Institutions.
egies
Write for our illustrated catalog with full descriptions of the different models.
Bausch ۤ lomb Optical G.
554 ST. PAUL STREET, ROCHESTER, N.Y.
NEW YORK WASHINGTON CHICAGO SAN FRANCISCO
Leading American Makers of Microscopes, Projection Apparatus, Photographic Lenses, Engineering Instruments, —
Stereo-Prism Binoculars, and other High-Grade Optical Products. :
SEER OT
—
ay
-
oe
-
"
+
UME XXV NUMBER 4
lies ‘THE °: ‘oil
JOURNAL OF ‘GEOLOGY
Lely Pal fees } celia
A SEMI- QUARTERLY
nasa),
EDITED By
THOMAS C. CHAMBERLIN AND ROLLIN D. SALISBURY
With the Active Collaboration of
SAMUEL W. WILLISTON, Vertebrate Paleontology ALBERT JOHANNSEN, Petrology
STUART EDIE Ds Invertebrate Paleontology ROLLIN T. GHEMBERLIN, Dynamic Geeleey y
i: Viet ALBERT D. BROKAW, Economic Geology
‘ASSOCIATE EDITORS
ARCHIBALD GEIKIE, Great Britain a JOSEPH P.IDDINGS, Washington, D.C. :
{ARLES BARROIS,,. France ; JOHN C.. BRANNER, Leland Stanford Junior University
RECHT PENCK, Germany f RICHARD A. F. PENROSE, Jr., Philadelphia, Pa.
NS REUSCH, Norway : WILLIAM B. CLARK, Johns Hopkins University
ARD DeEGEER, Sweden WILLIAM H. HOBBS, University of Michigan
V. EDGEWORTH DAVID, Australia — FRANK D. ADAMS, McGill University
LEY WILLIS, Leland Stanford Junior University ; CHARLES K. LEITH, University of Wisconsin
ROVE K. GILBERT, Washington, D.C. WALLACE W. ATWOOD, Harvard University
HARLES D. WALCOTT, Smithsonian Institution WILLIAM H, EMMONS, University of Minnesota
[ ENRY S. WILLIAMS, Cornell University R ARTHUR L. DAY, Sarees Institution
_ MAY-JUNE 1917
nag i Res Norte ay
IDOSAURUS COPE, A LOWER. PERMIAN oT Ges REPTILE FROM TEXAS ?
athe _ SAMUEL .W, WILLISTON 309
ES | ON THE 1916 ERUPTION. OF MAUNA LOA cai iaa = One ve Hey O. Wood = 322
i AND STRATIGRAPHIC RELATIONS OF THE OLENTANGY SHALE OF CEN-
-TRAL OHIO, WITH REMARKS ON THE PROUT LIMESTONE AND SO-CALLED
‘ OLENTANGY SHALES OF NORTHERN OHIO - - AMADEUS W. GRABAU 337
E HISTORY OF DEVILS LAKE, WISCONSIN - - - Arraur C. TrowsripcE 344
_ TAIN REGION - - - Seas - Eine NG = - €.W. TomMLInson 373
pay MeN era ah ical ronnie an cues Oh kata SH RG th ta Nhe rere BOS
Ere UNIVERSITY OF CHICAGO PRESS
CHICAGO, ILLINOIS, U.S.A.
i
; AGENTS:
THE CAMBRIDGE UNIVERSITY PRESS, Lonpon anp EpINBURGH
THE MARUZEN-KABUSHIKI- KAISHA, Tokyo, Osaka, Kyoto, FukuoKA, SENDAI
THE MISSION BOOK COMPANY, SHancuat
THE JOURR Tr GEOLOGY
EDITED BY
THOMAS C. CHAMBERLIN AND ROLLIN D. SALISBURY
With the Active Collaboration of
SAMUEL W. WILLISTON ALBERT JOHANNSEN
Vertebrate Paleontology Petrology
STUART WELLER ROLLIN T. CHAMBERLIN
Invertebrate Paleontology Dynamic Geology
ALBERT D. BROKAW
Economic Geology
The Journal of Geology is published by the University of Chicago at the University Press semi-
quarterly, on or about the following dates: February 1, March 15, May I, June 15, August 1, September 15,
November 1, December 15. ] The subscription price is $4.00 per year; the price of single copies is
65 cents. Orders for service of less than a half-year will be charged at the single-copy rate. {| Postage is —
prepaid by the publishers on all orders from the United States, Mexico, Cuba, Porto Rico, Panama Canal
Zone, Republic of Panama, Hawaiian Islands, Philippine Islands, Guam, Samoan Islands, Shanghai.
{| Postage is charged extra as follows: For Canada, 30 cents on annual subscriptions (total $4.30), on
single copies, 4 cents (total 69 cents); for all other countries in the Posta] Union, 53 cents on annual
subscriptions (total $4.53), on single copies, 11 cents (total 76 cents). ‘| Patrons-are requested to make
all remittances payable to the Universily of Chicago Press in postal or express money orders or bank
drafts. ©
The following agents have been appointed and are authorized to quote the prices indicated:
For the British Empire: The Cambridge University Press, Fetter Lane, London, E.C. 4. Yearly
subscriptions, including postage, 18s. 8d. each; single copies, including postage, 3s. 2d. each.
For Japan and Korea: The Maruzen-Kabushiki-Kaisha, 11 to 16 Nihonbashi Tori Sanchome,
Tokyo, Japan. Yearly subscriptions, including postage, Yen 9.00 each; single copies, including
postage, Yen 1.50 each.
For China: The Mission Book Company, 18 Peking Road, Shanghai. Yearly subscriptions,
$4.00; single copies, 65 cents, or their equivalents in Chinese money, Postage extra, if mailed direct
outside of Shanghai, on yearly subscriptions, 53 cents, on single copies, 11 cents.
Claims for missing numbers should be made within the month following the regular month of
publication. The publishers expect to supply missing numbers free only when losses have been sustained
in transit, and when the reserve stock will permit. Deliveries are not guaranteed within the war zones of
Europe, nor in other countries where transportation is disturbed by conditions incident to war.
Business correspondence should be addressed to The University of Chicago Press, Chicago, Ill.
Communications for the editors and manuscripts should be addressed to the Editors of THE
JOURNAL OF GEOLOGY, the University of Chicago, Chicago, Ill.
Contributors are requested to write scientific and proper names with particular care and in citations
to follow the form shown in the pages of the Journal.
Separates, if desired, must be ordered in advance of publication. Separates of leading articles will be
printed as ordered, of which 50 (without covers) will be furnished gratis, the actual cost of the remainder
(and covers, if desired) to be paid for by the author. The actual cost may vary, and will depend upon
the amount of work in re-making the pages into forms, presswork, paper. binding, etc. Separates
containing half-tones may be expected to cost somewhat more, the increase depending upon the number
of cuts and the amount of work required upon them.
Entered as second-class matter, March 20, 1893, at the Post-office at Chicago, IIl., under the Act of March 3, 1879
VOLUME XXV NUMBER 4
THE
foOuURNAL OF GEOLOGY
MAY-JUNE 1917
LABIDOSAURUS' COPE, A LOWER PERMIAN COTY-
LOSAUR REPTILE FROM TEXAS
SAMUEL W. WILLISTON
University of Chicago
Although remains of the genus Labidosaurus are not rare in the
upper Clear Fork beds of Texas, it has only been lately that fairly
complete skeletons have been secured. During the past season
Mr. Paul C. Miller was fortunate in finding, not far from the
paleontologically famous Craddock Ranch, near Seymour, Texas,
some half-dozen specimens, associated in a spot but a few square
yards in area—specimens which furnish nearly every detail of the
skeletal structure, excepting, as usual, the length of the tail. These
specimens come from a horizon that I have several times mentioned,
that from which the connected skeletons of Seymouria and Trimero-
rhachis, previously described, were obtained. Like those, these
skeletons were inclosed in hard clay nodules of a mottled red and
white color. The bones are rather soft and white in color, while
the matrix is very hard, making their preparation difficult.
t Labidosaurus Cope, Proc. Amer. Phil. Soc., 1896, p. 185; Case, Zodl. Bull.,
II (1899), 231 (Pariotichus); Revision of the Cotylosauria (1911), 45, 101; Permo-
carboniferous Beds of North America, and Their Vertebrate Fauna, p. 137, 1915; Broili,
Paleontographica, XI (1905), 51; Williston, Jour. Geol., XVI (1908), 359; zbid., XXII
(1914), 65; zbid., XXII (1914), 414; Contributions from Walker Museum, I, No. 8
(1914), 157; ibid., No.9 (1916), 221; Branson, Jour. Geol., XTX (1911), 136; v. Huene,
Bull. Amer. Mus. Nat. Hist., XXXII (1913), 351.
309
310 SAMUEL W. WILLISTON
The most important of these specimens (No. 174 Fig. 1), that
upon which the restoration (Fig. 8) chiefly is based, includes the right
half of the skull, the connected series of vertebrae and ribs to the
base of the tail, the pectoral girdle, the humerus, the radius, the
ulna, a single metacarpal, the pelvis, seen from below, and the
right hind leg complete, except the phalanges of the fourth and
fifth toes. Another specimen (No. 177) comprises the skull, a
Fic. 1.—Labidosaurus. Specimen No. 174, as prepared. One-fourth natural size.
complete series of closely articulated vertebrae and ribs to the sec-
ond sacral, the clavicular girdle in place, a part visible of the left
scapula, the left front leg, with the exception of most of the
phalanges, a femur, and part of the pelvis. This specimen has
been laid bare on the ventral side. A third specimen (No. 176)
has the skull with its upper part largely destroyed, the series of
vertebrae and ribs nearly to the sacrum, the pectoral girdle, both
humeri, one forearm, and part of the hand. The fourth speci-
men (No. 178) comprises a complete skull, the connected vertebrae
to about the middle of the back, and the connected pectoral girdle.
The other specimens are more fragmentary and have not yet been
prepared.
LABIDOSAURUS COPE 311
In addition to these specimens preserved in their matrix, there
are eight other more or less complete skulls, as many more incom-
plete, and numerous series of vertebrae, girdles, and limb bones—
altogether about thirty specimens. The material, it will be seen,
is ample to determine the skeletal structure.
In size, these specimens vary not a little. About half of them
are of nearly uniform size, the skull measuring seven inches along
the median line; these clearly all belong to the type species L.
hamatus Cope. Four other skulls, including the one originally
described by Case and myself as Labidosaurus (Pariotichus) incisivus
(No. 634) and Nos. 174 and 178 measure five inches along the mid-
line. All of these, unless it be No. 634, come from the upper Clear
Fork horizon. Two other skulls, and parts of others, including
the maxillae in which the additional maxillary teeth were observed,
are of smaller size, measuring only four inches. Two of these
(180 and 183) come from a much lower horizon, that which has
yielded most, if not all, the known specimens of Pantylus. They
have but a single row of teeth in the mandibles, and have the long
premaxillary teeth, and cannot be placed in the genus Captorhinus,
though they differ materially from the large skulls in having the
face less compressed. Probably they will eventually: find a place
in another yet unnamed genus. The most important of these is
a very perfect specimen (No. 183) herewith figured (Fig. 2). In
each orbit there is preserved a part of a sclerotic ring, the earliest
appearance, I believe, of such ossifications in a reptile. The separate
plates (Fig. 3) are narrow, with a flattened, thumblike projection
on each extending over the adjoining plate near its inner end; other-
wise they are not imbricated. In specimen No. 174 fragments of
similar plates are preserved in the orbit. Doubtless all the mem-
bers of this group had similar ossifications of the sclera, and it is not
at all improbable that most Paleozoic reptiles possessed them.
Skull.Nearly every detail of the skull of Labidosaurus is known
with assurance, owing to the combined researches of Cope, Broili,
Case, and myself. In the first figure of the skull given by me, the
quadratojugal was given as a distinct bone, and also a division of
the squamosal into two bones. In my second figure the quadrato-
jugal was omitted as doubtful, as was also the supratemporal. I
312 SAMUEL W. WILLISTON
thought that I found in two specimens a suture separating the
posterior part of the squamosal, and recognized it as the “‘epiotic,”’
following Cope. A skull acquired very soon thereafter showed
clearly that there was no such suture, and that the ascription of a
tabular was an error, as recorded by Branson (op. cit.). The
sutures given in the present figures have been corroborated in at
least a dozen skulls, and there can no longer be any doubt about
Fic. 2.—Labidosaurus(?). Skull, from the side. Three-fourths natural size.
No. 183.
them. ‘The figure given by v. Huene (oP. cit.), I regret to say, has
a number of inaccuracies. There is a distinct quadrotojugal, a
bone found in nearly all the American Permocarboniferous reptiles;
there is no supratemporal, absent also in Procolophon and Cap-
torhinus; and no tabular, present in all other cotylosaurs, and
in many theromorphs. Of the structure of the under side of
the skull I have no corrections to make of my earlier figures. In
none of the large specimens is there a trace of the parasphenoid;
it is certainly absent in some; in two of the smaller skulls it is
present though small. Perhaps its absence.is a generic or a specific
character. Nor have I any changes to make in my figures of the
LABIDOSAURUS COPE 313
occipital region. The paroccipital exists as a distinct bone. The
prootic, epipterygoid, and the relations of the quadrate are as I
have described and figured them. The small bones above the brain
case, which I doubtfully called the “‘alisphenoids,”’ in the doubt
now existing as to the presence of the mammalian alisphenoid in
the reptiles may be called the “postoptics,”’ the name first pro-
posed for them by Cope. Nor have I any emendations to make of
the structure of the mandible as
figured by me.
Whether all the forms have
additional teeth on the inner side
of the maxillae posteriorly I can-
not say since in most specimens
the mandibles are tightly closed,
precluding an examination of
this region. Two smaller max-
illae, as recorded by Branson,
possess them. The premaxilla
has, in most specimens, three
teeth, of which the first is the largest, and the third small. Case has
based a species, L. broiliz, on the presence of two elongated premax-
illary teeth, but I think that the character is variable. The number
of teeth in the maxilla also is variable. In the earlier specimens I
could find not more than seventeen or eighteen. In that herewith
figured (No. 174) there are twenty-four, the posterior ones quite
small. Whether this difference is a specific or an individual
character I am not prepared to say. Iam unwilling to give specific
names until I am assured that I know specific characters, and such
we do not know well in any genus from the American Permian—
in general I am skeptical of the ‘‘species”’ of all fossil reptiles.
Nearly every skull of Labidosaurus hitherto found was fossilized
in a horizontal position and it has been exceedingly difficult to
determine the amount of depression they had suffered. Among the
numerous skulls now at my command there are two only which have
not thus been distorted. One, an unusually large skull measuring
eight inches in length, shows very little if any distortion. The
other, that herewith figured (Fig. 1), had been evenly split along the
- Fic. 3.—Labidosaurus. Sclerotic
plates. Enlarged.
314 SAMUEL W. WILLISTON
median line and the right half has been preserved in articulation
with the vertebrae lying upon its side. In this specimen, the slight
distortion is in the other direction; its height posteriorly is a little
too great. From a study of all these specimens, however, one can
determine almost exactly the elevation of the posterior part of the
skull, about as I have figured it in the restoration. Whence it
follows that in all the figures hitherto given, including my own, the
skull is shown too much flattened posteriorly, and the orbits too
Fic. 4.—Labidosaurus. Anterior dorsal vertebrae. A, from the side; B, from
in front; C, from below; D, from above. Natural size.
-ovate in form when seen from the side. In reality the orbits
seen from the side are almost circular in outline.
Vertebrae.—The peculiar structure of the vertebrae, which
Labidosaurus shares for the most part with Captorhinus, has been
described sufficiently well by Case, and will be seen from the
accompanying figure (Fig. 4). Vertebrae of this type are known
exclusively from the Texas deposits, with one exception, a specimen
of a large form found by Professor Case in the El] Cobre Cafion.
I believe that when the skull of this genus is discovered it will prove
to be of the Labidosaurus type. I may add that the horizon of
this specimen is rather high up in the Cobre deposits.
LABIDOSAURUS COPE 315
The atlas cannot be made out in any of the connected specimens.
Doubtless there is also a proatlas. The axis scarcely differs from
the following vertebrae. There are twenty-five presacral vertebrae,
definitely shown in specimen No. 177. ‘This is the number found
by Case in Captorhinus, and one less than the number in Limno-
scelis, a genus which has not a few points of resemblance to
Labidosaurus. It is two more than is possessed by Seymouria,
and three or four more than in Diadectes. In Pantylus the num-
ber is unknown.
The postaxial, presacral vertebrae are scarcely distinguishable
from each other, except the five posterior ones, which have no pro-
cesses for rib articulation. The spines anteriorly are a little more
slender, giving greater freedom of vertical movement. The trans-
verse processes anteriorly are a little longer, standing out beyond
the margin of the zygapophyses; they are also a little stouter here,
extending down farther on the sides of the centrum. The inter-
centra anteriorly are small, not nearly filling out the space between
the margins of the centra. There are two sacral vertebrae; the
first bears a stout rib, with a broad, vertical face, as long as the
centrum, for articulation with the iltum; the second has a much
smaller rib, only touching the illum behind and below the first rib.
The caudal vertebrae, so far as known, are narrower than the pre-
sacral, and have more slender processes. ‘Their ribs are co-ossified
with the centra, unlike those of Seymouria.
The first and second ribs are slender; the third is broader and
much longer. The fourth, as in Captorhinus, is remarkable (Fig. 1)
and would hardly be recognized as a rib if found isolated. Its
length is less than three times the width of the distal end. It is
narrowed in the middle and much expanded at its extremities.
Like the early ribs of Seymouria, Diadectes, and Limnoscelis, its
function was the support of the scapula.
Pectoral girdle (Figs. 5, 6).—In five different specimens, the pec-
toral girdle, more or less complete, is found in position with the
skull and but slightly dislocated from its association with the
vertebral column. In each case the front end of the interclavicle
with its articulated clavicles lies between the angles of the jaws
and immediately under the occipital condyle—precisely its posi-
210 SAMUEL W. WILLISTON
tion in the temnospondylous amphibians and in Dzplocaulus.
There was literally no neck in the cotylosaurs. In three of these
specimens the shaft of the clavicles is turned dorsad at a right
angle to the lower, horizontal, and expanded part; in two or three
others the angle is slightly greater. In one specimen, that shown
in Fig. 1, lying in immediate connection with the upper end of the
Fic. 5.—Labidosaurus. A, front leg; B, hind leg; C, clavicular girdle clavicles
from below, interclavicle from below, clavicles from in front; D, humerus ventral
face; E, scapula, outer face (No. 634). All figures about one-half natural size.
clavicle, there is a very slender bone about 20mm. in length; a
similar bone is also found in specimen No. 178 with quite the same
relations. There would seem to be little doubt that it is a vestigial
cleithrum. In Limnoscelis the cleithrum is very small, though
much larger in Diadectes: None has been observed in Seymourza,
but I am inclined to believe that all the American cotylosaurs
possessed the element, though in some cases it was vestigial. The
scapula figured (Fig. 5, £), is quite complete, but is lying on its
LABIDOSAURUS COPE 317
side and has suffered a lateral compression. In specimen No. 178
(Fig. 6) the scapulae lie almost perfectly in position with the
clavicular girdle, or at least that of the left side. The blade may
have been pressed outwardly a little, but the angle between it and
the horizontal coracoids is nearly rectangular. As a whole, the
scapula is rather broad and short, as in Limnoscelzs, a little narrower
in the larger specimen. The sutures separating the coracoids and
the scapula have been corroborated in several specimens.
Fic. 6.—Labidosaurus. No. 178. Pectoral girdle, from below. One-half
natural size.
Forelegs.—The humeri of the different specimens vary appre-
ciably in shape, but I do not know the value of the differences.
They have a greatly expanded distal end, as would be called for
by the large and stout hand. The radius and ulna are shown con-
nected in three specimens. They are somewhat more slender than
the posterior epipodials, and are much shorter than the humerus.
The carpus was correctly figured by me in my earliest paper;
unfortunately some of the bones were incorrectly identified: it is
the fifth and not the first digit which was missing. ‘There are
definitely four phalanges in the third finger, and there can be little,
if any, doubt that the formula was the primitive one of 2, 3, 4, 5, 3.
The fifth carpale I have not recognized, but it doubtless was ossified.
318 SAMUEL W. WILLISTON
Pelvis —Of the pelvis the ilium is not visible in its entirety
in any specimen. In the restoration I have borrowed from Broili’s
figure the outline of the upper part articulating with the sacral
ribs. The flat, platelike pubes and ischia are firmly united in the
middle, with no opening except the usual obturator foramen; they
are proportionally long.
Hind legs—The femur has been well figured by Case. It is
rather short and stout, much stouter than in Captorhinus. The
correctly figured by Case. In one
specimen (No. 174) I find what I
believe to be evidence of two cen-
tralia, as in Ophiacodon, though
Case figures but one. The phalan-
geal formula was doubtless 2, 3, 4,
Fic. 7.—Labidosaurus. Outline of 5» 4 The first three digits are
skull, from above: pm, premaxilla; shown in the figure (Fig. 1). The
n, nasal; m, maxilla; /, lacrimal; first digit of both the hand and the
Sree ieee a As an foot is relatively large, and its
pa, parietal; sq, squamosal; ds, der- metapodial was capable of but
mosupraoccipital; pf, parietal fora- little divarication from the others.
Pen ae trom Water Rep- The fifth metapodial was but little
se Ts shorter than the fourth.
Parenthetically I may add that I do not at all agree with Good-
richt in imputing to the form and position of the fifth toe so much
importance in the classification and phylogeny of the reptiles.
The divarication and hook shape of the fifth metatarsal have been
due, I believe, to modifications of the tarsus and doubtless have
arisen homoplastically in various lines of descent. In all primitive
reptiles there was a fifth tarsale. As has been amply proved, I
t Pyoc. Royal Soc., LXXXIX (1916), 261.
Ea tibia is unusually stout, and the
( fibula is much curved; both bones
are much shorter than the femur.
My original figure of the tarsus
* was erroneous in several particu-
; lars, as was suggested by Jaekel.
f Most of the tarsal bones have been
7
LABIDOSAURUS COPE
think, the bone has absolutely disap-
peared in all modern amniotes, leav-
ing a space that was soon occupied
by the proximal end of the fifth
metatarsal, which came to articulate
directly with the large fourth tarsale.
In many Permian reptiles a divarica-
tion and proximal elongation of the
fifth metatarsal had begun. This
change in the articulation, it seems
to me, will easily account for the
change in the form of the fifth meta-
tarsal in crawling plantigrade rep-
tiles. In the rectigrade, and more
ambulatory reptiles, on the other
hand, with less functional use of the
fifth toe, the metatarsal retained more
of its primitive shape and position
parallel with the fourth, especially in
the forms ancestral to the mammals.
The chief fallacy in Goodrich’s
arguments lies in deriving the stego-
crotaphous chelonian skull from a
single- or double-arched ancestor
because of the shape of the fifth
metatarsal. It is now conceded that
the turtles must have had a direct
ancestry from the cotylosaurian type
of reptile; if so the hook-shaped
metatarsal must have been an inde-
pendent acquirement.
The hands and feet of Labidosau-
rus (Fig. 5) are relatively large and
powerful, the foot nearly as long as
the leg, the hand as the arm. The
first digit of each is relatively large
and but little divaricated in life; its
3219
About one-fourth natural size
Skeleton.
Fic. 8.—Labidosaurus.
320 SAMUEL W. WILLISTON
metapodial is but little smaller than the second. The metapodials
are all stout, but the digits are short, tapering rapidly. The ungual
phalanges are more pointed than in either Limnoscelis or Diadectes;
the fifth tarsale, as usual, is small.
Very slender ventral ribs, for the first time recognized in this
genus among the cotylosaurs, are present In numerous specimens.
So far as they are preserved, they do not seem to differ from those
of Ophiacodon or Varanops, for instance. It is pretty certain that
they are absent in the diadectids and Limnoscelis. Why some
genera in each order of reptiles should have retained these bones,
while others lost them, I am not prepared to hazard a guess.
Possibly they have something to do with water habits.
Fic. 9.—Labidosaurus. Life restoration, about one-eighth natural size
Restoration (Figs. 8 and 9).—The only attempt that has hitherto
been made at the restoration of the skeleton of Labidosaurus is that
of Broili, which was quite acceptable, considering our little knowl-
edge of the genus and its allies at the time he made it. That he
placed the front legs so far back as he did, is not surprising; almost
anyone would have done likewise at the time; and in the curvature
of the vertebral column he was unduly influenced by the erroneous
early restorations of Pareiasaurus, at that time supposed to be a
near relative. I may only remark here that the present restoration,
based as it is upon so much material, has scarcely anything con-
jectural about it, or anything that has not been corroborated by
several specimens. As usual, the length of the tail is still in doubt.
I have assumed that it was about as long as in the allied captor-
hinids, of which we know the tail more definitely.
Habits —Granted a fairly complete knowledge of the osseous
structure of such early reptiles, there must remain more or less
LABIDOSAURUS COPE 321
conjecture as to how and where the creatures lived. The two
striking characters seen in the skeleton are the elongated, hook-
like incisor teeth and the very powerful feet. Were they corre-
lated? I believe that they were. There is no evidence that the
powerful feet were developed for use in swimming, though doubt-
less the creatures swam well. Nor is there any evidence that the
creatures used them for digging holes in which to crawl, for the
very good reason that the front legs were altogether too short;
they could not have scratched their own noses, much less dig
holes for the body to enter. Why not then assume that the feet
were developed for digging for food? The epipodials are relatively
short, a character constantly found in animals using their legs for
propulsion in the water. But why not the same shortening to give
greater immediate power in digging? I believe, then, that
Labidosaurus lived about the lowlands, on the borders of the seas
or lakes, and found its livelihood in extricating worms and larvae
from the rocks or soil, for which the long, hooklike front teeth were
admirably fitted. The posterior teeth in Labidosaurus were not
strictly carnivorous; they were better adapted for cutting than for
tearing or seizing. They are flattened, with an obtusely pointed
apex; and the palatine teeth were very small, except those of the
“transpalatine” part of the pterygoids. The motion of the head
upon the shoulders was limited laterally, rather free vertically.
The trunk was not very flexible. The spines were short, but the
great development of the arches of the vertebrae furnished ample
place for the attachment of muscles controlling the head. Finally,
in motion the animal must have been slow and sluggish.
In some respects Labidosaurus is the most specialized of all
the American cotylosaurs, especially in the loss of the supra-
temporal and tabular bones, rarely absent in other cotylosaurs;
and in the reduction of the dermosupraoccipitals, and their restric-
tion to the occipital surface of the skull. The genus could not have
been ancestral to any known later cotylosaurs, though possibly
the Captorhinidae may have been allied to the ancestral stock of
the Procolophonia of the Trias.
NOTES ON THE 1916 ERUPTION OF MAUNA LOA
HARRY O. WOOD
Hawaiian Volcano Observatory
The writer’s observations and comments on this eruption divide
naturally into four parts:
I. Distant observation and photographic record of the out-
bursts of fumes on May 19, and of the beginning of flow on
May 21-22, 1916.
II. Observation at the front of the Honomalino branch of flow
on May 23. ,
III. A hurried reconnaissance of the Kahuku branch of flow,
and of the flow-source region, on May 30-31.
IV. A thorough examination and photographic record of con-
ditions throughout the region of the source of flow, made in the
company of Dr. A. L. Day, in the six days, June 28—July 3, 1916.
Treatment under these headings makes for a somewhat extended
and rambling account; but it has been found very difficult to
present the complicated sequence of observations, which in some
respects are unrelated, in any more succinct way.
LE
The beginnings of this eruption were noticed first in the early
morning hours of May 19, 1916. No immediate premonitory
symptoms were recognized previously. The earliest observations
which have come to notice were made by Captain D. F. Nicholson,
of the steamship ‘‘Hamakua,” and by Mrs. R. A. McWayne, of -
Papa.
At about 3:45 in the morning, while sitting on the bridge as
the ‘‘Hamakua’’ was steaming around South Point, Captain
Nicholson experienced an earthquake. The sea was smooth and
« Much of the matter in Part I was published at once in the Weekly Bulletin of
the Hawaiian Volcano Observatory, IV, No. 5, pp. 34-37; but this report was neces-
sarily fragmentary, and partly erroneous, so it is restated here.
322
NOTES ON THE 1916 ERUPTION OF MAUNA LOA 323
the weather calm and cloudless. Suddenly the ship trembled from
stem to stern, as though it had grounded on a beach, and loose
things in the ship’s galley were disturbed; also the smooth sea
surface suddenly became agitated violently by the commingling of
several different systems of waves, an action which kept it in a
state of turmoil for about a minute. (No shock so strong as this
one was recognizably registered in the Whitney Laboratory of
Seismology at Kilauea.)
A little later, at about 4:15 A.M., Captain Nicholson heard a
sound lasting for about three seconds, which he likened to a volley
of musketry, and he then saw what he described as a spiral column
of fumes rising from a point high up on the south flank of
the mountain. From her residence above the road near Papa,
Mrs. McWayne in the early morning hours saw a bright glow high
up the mountain.
During the morning and forenoon hours of May 19 a swarm of
local earthquakes was registered at the Whitney Laboratory. All
these were extremely feeble shocks, even when considered from
the seismographic point of view. ‘The earliest of them was recorded
a little before three in the morning. Beginning in the late forenoon
a lull followed, less than twenty-four hours in duration.
Throughout the evening and night of May 18, and the morning
and forenoon of May 19, the weather was brilliantly clear. Looking
westward from the Hawaiian Volcano Observatory, situated on the
northeast margin of the crater of Kilauea, one could see nearly the
whole profile of Mauna Loa outlined sharply against the sky.
From late evening until dawn there was brilliant moonlight.
Until midnight, and at 6:15 A.M., no signs of eruption were seen
from the vicinity of the observatory. However, at about 6:15 A.M.
the beginning of a fresh outburst of fumes, high on the mountain
slope, was seen by Captain Nicholson, whose ship had then come
to anchor at Honuapo.
At about 7 A.M., perhaps a little before, a definite outbreak and
uprush of fumes became visible at Kilauea. At first a group of
cloud-forms appeared high up on the south flank of Loa, rising
from behind the mountain profile at a distance of at least twenty-
five miles, as viewed from near the observatory. Here these were
324 HARRY O. WOOD
first seen by Joseph Moniz. Though their appearance and develop-
ment exhibited peculiarities, Moniz at first thought them to be
ordinary cumulus clouds, such as frequently rise from behind the
mountain. Yet they held his attention. After about ten minutes,
since they continued to rise straight upward at the same place, he
pointed them out to others; but still there was doubt as to their
character, and he allowed more than a half-hour to pass by before
he called the attention of the writer to them, at about 7:35 A.M.,
having grown surer that they were columns of fumes.
When first seen by the writer there were two chief standing
columns in which fumes were swirling rapidly straight upward and
merging in a double cumulus crown. ‘The column higher up the
slope was somewhat larger and taller than that lower down; and
these were then separated by a clear interval whose width was
about the same as the diameter of the larger column—about 1,000
feet. As soon as possible the writer began making a series of
photographic records of this action, illustrating its development
throughout the rest of its duration.t_ Five of these are shown here,
Plate II,a, b,c,d; Plate III,a. The life of this outbreak was short.
The higher fume column then rose above the mountain profile
probably from 11,000 to 12,000 feet, and the lower from 8,000 to
10,000 feet. Very quickly other subordinate columns appeared,
first at both sides and then in between the chief ones, and soon all
had merged into a single pillar of uprushing fumes, issuing more
and more copiously and rising to higher and higher altitudes. By
7:45 (Plate II, a) the diameter of this column had thus increased
to more than a mile, where a little earlier the total width including
the clear space did not exceed 3,000 feet. The column had now
reached a height of 15,000-18,000 feet above the mountain profile.
Its stem resembled a huge column fluted with drapery hung in
simple vertical folds. The cumulus crown still showed a double
head, and thus continued to indicate the positions of the two
dominant fume columns which, in reality, persisted throughout
t These views—12 (14) in all—were made on Wratten Panchromatic M plates,
quarter-plate size, through a K; filter and a Zeiss Double Protar lens, F=13 cm.,
with a stop of f45.. Time exposures of about five seconds were made with the earlier
views; but, with increasing actinicity, the later exposures were clipped a little short.
NOTES ON THE 1916 ERUPTION OF MAUNA LOA 325
the outbreak. Also a smoke ring is shown encircling the upper
part of the column below the crown. By eye observation the
writer did not notice this ring until about 7:50 a.m. Despite this
he feels confident that it had not begun to form when he first saw
the eruption cloud.
After 7:50 A.M. the form of the eruption cloud underwent rapid
changes as the continued emanation of fumes added to its bulk,
and convection currents and varying winds at different altitudes
continually reshaped it. At 8:00 A.M. the diameter of the stem
had increased to from 13 to 13 miles and the smoke ring, which had
rapidly enlarged, had begun to fray and spread horizontally about
equally in all directions, except for a slight elongation toward the
northwest, forming the striking mushroom shown in Plate IT, 0.
By this time the crown had reached a height of 20,000 feet, or
more, above the mountain profile, and its tip was just beginning to
fray in the upper wind.
At a little after 8:00 A.M. the uprush of fumes began to dimish,
and by 8:30 the two dominant columns were again separated, and
the subordinate columns had ceased to rise continuously. The
fume cloud had spread rapidly at the level of the smoke ring,
forming a cloud blanket, and this, with the fraying and drift to
eastward from the summit of the crown, gave rise to the beautiful
cloud-form shown in Plate II, c. This has been likened aptly to a
ballet dancer. The emanation of fumes continued to diminish
rapidly, as is emphasized in the view (Plate II, d) taken at 9:05 A.M.
Only a graceful cloud-form then remained, with thin columns of
rising fumes. ‘
By 1o A.M. (Plate III, a) it required keen observation to detect
any further output of fumes; and by 10:30 this could be made out
only by experienced eyes. By 11:00 A.M. the action was doubtful,
and it grew more and more doubtful afterward—though cloud-
forms occasionally appeared where undoubted fume columns had
been rising. By noon nothing could be seen but a frayed stratum
of high cloud overspreading the sky above the mountain. This
- exhibited ripple-marking, like cross-waved cirrus. It persisted
until after sunset.
326 HARRY O. WOOD
From about 8:30 A.M. on, short-lived, subordinate columns,
wisplike in appearance, were noted at points a little higher up the
slope than previously and at points much farther down the slope
also, over a span from five to seven times as great as the width of
the fume column at its greatest. These lateral columns did not
persist individually, and gradually they ceased to appear.
Until the smoke began to fray and spread out like a blanket,
the columns of upcurling fumes were fleecy white in appearance
in the bright morning sun, with cream tints also, like cumulus
cloud; and so too was the cumulus crown. As the blanket spread,
however, it shadowed first the column and then its own lower
surface, so that these shaded portions took on a faint, graduated
coloring in which brownish and purplish tones of a faded-out
quality were commingled with various shades of gray. This
coloration developed quickly with the horizontal outspread of the
fumes. No truly dark-colored fumes were seen.
After the cumulus crown had risen into the upper air and had
begun to fray and drift eastward, such action continued until the
lessened emanation of the fumes brought it to an end, late in
the forenoon. Altogether, however, only a small percentage of the
fumes reached the uppermost levels. Most of the drift was to the
northwestward. By 8:00 a.m. this tendency for the horizontal
fume-cloud blanket to draw out in the northwest direction was
noted. By 9:30 the drift in this direction of the blanket as a whole
had become noticeable, and by 10:30 such shifting was marked.
This spreading and drift of the cloud blanket to the west and
northwest until it stretched out back of the mountain profile, so as
to extend below the skyline, made a cloud-colored background
against which were rising the cloud-colored fumes, slightly tinted
with brown. This made the fuming very hard to see, but there is
no doubt of its rapid and progressive diminution after 8:00 A.M.
During the afternoon no rising fumes were seen definitely. But
for a very short time, just at sunset and after, very thin, translucent
fumes were seen, brown in the transmitted light of the western sky.
In the early evening there was cloud and mist intermittent with
brief, clear glimpses. From the observatory no definite glow
could be seen, but there appeared to be a very faint radiation of
NOTES ON THE 1916 ERUPTION OF MAUNA LOA 327
light over the dark sky from a point on the slope near the place of
outbreak, though apparently a little lower down. That this
emanated from the eruption was, and is, doubtful. It was too
faint to be a positive illumination. However, it is the under-
standing of the writer that a faint glow was seen that night by
Captain Nicholson when off Fisherman Point. This and subse-
quent events strengthen the probability that the faint light seen
from the observatory was, in reality, from the eruption.
From the observatory no fumes were seen on the 2oth, nor
early on the 21st. From Kealakeakua, however, a “‘pillar of
smoke”’ was seen early in the morning of the zoth by Miss Paris,
a lifelong resident of Hawaii, familiar with the appearance of the
mountain profile and with the characteristics of eruption here.
This appeared high up on the south slope as seen from Kona.
From the observatory, just at sunset in the evening of the 21st, thin
brown-toned fumes were seen by transmitted light rising from near
the place of outbreak. They appeared somewhat more pronounced,
even, than on the evening of the roth.
A small amount of lava probably was ejected at the time of this
outbreak. This was reported as seen by men high on the slopes
back of Naalehu and Kahuku, and in Kona, but the action was
quickly over.
As mentioned earlier, the swarm of earthquakes which preceded
and accompanied the first uprush of fumes was followed by a lull
in the registration of shocks, of less than twenty-four hours’
duration. After this they began to register in greater number
than before, and in most instances with greater amplitude also.
Intervals of quiet were short. The resumption and continuation
of this seismic activity led to expectation of the outbreak of flow.
Flow broke out, considerably lower down the slope than the place
of first outbreak, in the late evening of May 21. With little doubt
it was first seen by the writer, at about 11:15 P.M. It was at once
brought to the attention of others at the Volcano House near by,
and from here the news was spread by telephone on the eastern side
of the island. On the western side of the island the outbreak was
first noticed a little before midnight by Mrs. McWayne, and there
the news was similarly spread.
328 HARRY O. WOOD
Probably the outbreak was noticed by the writer almost as soon
as it occurred. The night was clear, except for a low bank of
clouds at the northeast. At 10:00 P.M. and earlier there was no
suggestion of illumination anywhere along the southern segment
of the mountain profile. At about 11:15 p.m. the writer went
from the Volcano House to the observatory to see that the seis-
mographs were in good working order before retiring (for, owing
to the frequency and energy of the shocks, there was likelihood of
the disarrangement of the struts and levers). The moon had just
risen, but it was still hidden behind the bank of clouds at the
northeast. However, a faint, diffuse moonlight spread over the
sky. Upon mounting the observatory porch a very faint light was
seen radiating from a point behind the mountain profile much lower
down the southern slope than the place of first outbreak. The
effect was similar to that sometimes produced here just as a planet
or a bright star drops below the mountain profile. After inspection
of the seismographs, however—perhaps three minutes later—this
radiating light had become more definite, although the illumination
of the sky by the moon had grown brighter. Almost at once it
took on a pinkish hue. Then the word was spread instantly.
Judging by the rapidity of its development in the first ten minutes
after it was discovered, the outbreak did not take place (or
more strictly, become visible from the observatory) earlier than
11:00 P.M., and probably not earlier than 11:10 P.M. Its discovery
was little later than its occurrence.
During the ten or fifteen minutes after it was first seen the
light at the fountainhead grew rapidly, and a small diffuse cloud
of fumes appeared. This increased quickly also, and the glow
steadily grew brighter and more ruddy. Soon the fumes at the top
began to drift to the northwest up the mountain slope behind the
profile. At the same time the progressive extension of a faint
ruddy illumination down the slope behind the profile was detected.
Flow had begun. As soon as possible, at about 11:45 P.M., the
director of the observatory, Professor T. A. Jaggar, Jr., and the
writer set out by motor and proceeded southwestward and west-
ward to a point near the boundary of the Kona district of Hawaii
beyond the western branch of the flow of 1907—a distance of about
NOTES ON THE 1916 ERUPTION OF MAUNA LOA 3290
sixty miles—and back again, arriving at the observatory at about
6:45 A.M., May 22. West of the village of Waiohinu several stops
were made, both going and returning, to observe and photograph.
All along the way from the observatory to the turning-point, and
back again, a gradual and steady increase was noted in the height,
amount, and spread of the fumes; and, until dawn, in the brilliancy
of the illumination at the fountainhead. A well-defined northwest
drift of the fumes in the upper strata of the air gradually developed.
From the upland flats along the road near Kahuku, and points
to the westward, a long line of faint reddish illumination was seen
extending to the right from the fountainhead. The course of this
was judged to be about south-southeast from the source. At first
it was considered to be light from the surface of a pool, but as it
elongated rapidly it was soon thought to be a line of illumination
above a flowing stream. This opinion was confirmed by a visit to
this flow made later in the day by Messrs.. J. W. Waldron and
T. Hardy.
At about 4:00 A.M., May 22, we met Mr. Samuel Kauhane at
the roadside gate of the ranch house at Kahuku—a man to whom
the south slope of Loa was well known—and he expressed the
opinion that the outbreak was higher up the mountain than the
group of old cones at Puu o Keokeo. This proved to be the case.
Upon our return to the observatory a photograph (Plate ITI, 0)
was made at 8:30 A.M. (as early as weather conditions would
permit), to show the position and development of the fume column
and crown rising above the fountainhead. This view should be
compared with Plate II, a to gain an understanding of how much
lower down the mountain than the place of earlier outbreak the
place of later outbreak is situated. The true azimuth from the
observatory of the apparent center of this fume column at the
source was found to be about S. 66° W. (the azimuths of the upper
and lower limits of the greater column of the earlier outbursts
were approximately S. 82° 30’ W. and S. 85° 30’ W.). This azimuth,
projected upon the government map, indicated a source low on the
southwest flank of the mountain, and, assuming this source to lie
in the line of the south-southwest rifting from summit to sea, it
was near the line of the upper branch of the flow of 1907 at an
~ 330 HARRY O. WOOD
altitude of 6,500 feet—as shown approximately on the government
map—a little above Puu o Keokeo. This location was confirmed
by field survey, though multiple mouths were found along a linear
rift at the source, and the altitude was found to be a little higher
than 6,500 feet at the lowest point of the actual source. The
region of the fountainhead thus is between 30 and 35 miles west-
southwest from the observatory.
_ In contrast to this, the region of the earlier outburst, deter-
mined by projecting the azimuths given above, is intercepted along
the course of the great rift-line between approximate contour lines
drawn on the government map, as follows: the upper and lower
limits of the great trunk column of uprushing fumes are thus
indicated at about 11,600 and 11,000 feet above sea, and the
diameter of the column is thus indicated at considerably over a
mile; similarly, the approximate upper and lower limits of the span
marked by subordinate sporadic columns of rising fumes may be
taken as 12,000-++ and 10,o0o— feet, determining the width of this
at five miles, or more. All these are approximate values. Never-
theless, this source is thus found to stand higher up the mountain
than the early estimates placed it.
In the late afternoon of the first day of eruption, May 22, the
writer returned to the southern part of the island, and spent the
evening and night in observing the changes in the magnificent
illumination from places along the road over the upland flats west
of Kahuku. This locality was reached just at nightfall; it was
then cloudy, with brief showers; however, before long the clouds
lifted, though they continued to cover the sky. Little by little
the character and extent of the illumination became visible.
Since the dark hours of the morning a great change had taken
place. At the fountainhead both the action and the illumination
were somewhat greater than when last seen in the early morning,
but here the change was least. The faint red illumination extending
toward the right (toward Waiohinu over upland Kahuku) had died
out. But toward the left, toward Kona, a long line of brilliantly
illuminated fume and cloud demarked the course of a flow which
had rushed down the mountain toward Honomalino. In early
evening this was still advancing at a considerable rate. The
NOTES ON THE 1916 ERUPTION OF MAUNA LOA B31
marked illumination, which we may designate as the primary glow,
was most brilliant at the fountainhead, and above the front of this
flow (where its outflashing was augmented by the light from the
burning forest); but also above the course of the flow between its
front and source the glow was much brighter than the general sky
glare. This formed a band of primary illumination whose length,
as seen from the gate about a mile west of the ranch house at
Kahuku, was very close to seven miles; and its brilliantly lighted
arch rose about three-quarters of a mile above the mountain profile
(Plate ITI, c).
A diffuse red glow covered the sky everywhere and, in early
evening, low-lying cloud and fog banks clinging to the mountain
slopes below the road, and illumined dimly by reflections from the
cloud layer above, led to a current erroneous opinion that the
flow already had advanced down the slope beyond the road in
Kona.
Owing to the wretched state of the road surface, a serious con-
gestion of motor traffic (some 250 motors headed westward and
about 80 headed southward, toward a meeting-point near Hono-
malino, on a road too narrow for passing or turning except at
widely separated places), and much uncertainty as to the exact
course, rate, and behavior of the flow, the writer spent most of the
night at a gate on the road about a mile west of Kahuku, where
he returned after going to within three to four miles of a group
of houses at Honomalino which stood in the apparent path of
the flow.
The photograph (Plate ITI, c') shows the scene from this station
as it appeared just before midnight, May 22. In this view, of
course, the bright reddish glow which covered all the sky does not
appear, but simply the brilliant arch of the primary illumination,
as designated above. During the night the following changes were
observed:
A little after midnight there was noticed a rapid spread of very
brilliant glow to the right of the fume column rising at the fountain-
head. This was a conspicuous feature of the action until after
1:00A.M. It was thought to indicate a flow toward Kahuku. This
1 Made on a Wratten Panchromatic M plate exposed 30 minutes at f/6. 3.
332 HARRY O. WOOD
judgment was confirmed by Messrs. Waldron and Hardy, who
witnessed its outrush from the high camp they occupied that night.
Beginning gradually, probably before midnight, certainly as
early as 1:00 A.M., there was noted a rapid decrease’ in the brillancy
of illumination above the line of flow extending toward Honomalino.
By 3:00 A.M. this, as seen from our station, was quenched com-
pletely; there remained only the diffuse glow of the clouded sky
and the brilliantly lighted column of fumes rising at the fountain-
head, and a much subdued glow above the front of the flow. And
this last was decreasing rapidly. By 4:00 A.M. the earliest light of
dawn found the illumination at the fountainhead much like that of
the previous morning, with the lines of illumination above the
courses of flow almost wholly quenched.
Shocks of earthquake continued to occur. Some were strong
enough to disarrange the struts and levers of the seismographs.
This made it inadvisable for the writer to be absent from the
observatory except at times when the director could be present;
so, after a brief visit to the front of the Honomalino branch, the
writer returned to Kilauea.
During the evening and night of May 23-24 it was seen from
the observatory that the glow had extended far to the left of the
fountainhead in a direction estimated at south-southeast. No
such extension of the band of illumination had been seen in the
evening or night of May 22-23. This was due to the renewal of
flow toward Kahuku on a much larger scale than in the beginning.
Ultimately the Kahuku branches developed much greater magni-
tude than those in Kona. During this evening and _ night,
May 23-24, this glow was seen through shifting clouds, so that no
good opportunity for making a photographic record presented
itself in the earlier hours. And, owing to his complete loss of
sleep on the two previous nights, the writer undertook no prolonged
watch. During the evening and night of May 24-25 this illumina-
tion appeared elongated farther to the south, and perhaps slightly
abated in intensity. A photographic record of this (Plate ITI, d)
«It should be noted that moonrise occurred between midnight and one o’clock,
but that this decrease in illumination was positive, nevertheless, as evidenced by the
glow above the Kahuku tongue.
NOTES ON THE 1916 ERUPTION OF MAUNA LOA 333
was obtained in the early morning hours of May 25, from 1:15 to
1:45 A.M. Ina lessening degree this glow was seen in the evenings
of May 25 and 26, and very faintly on the evenings of May 27
and 28. On the latter date it was near to the vanishing point.
II
The writer reached the front of the Honomalino branch of this
flow at about 11:00 A.M., May 23, at a point about three miles, by
trail, above the road. Here the flow was of a—a, still advancing at
a slow rate of speed. This was difficult to estimate on account of
the irregular character of the ground and the brief time available
for watching. Though moving much more slowly than earlier,
the advance was still steady. Possibly the maximum forward
movement of any considerable section of the front was ten feet in
an hour. The average over the whole front was less, perhaps four
to five feet in an hour. At this front the flow was narrow, not
more than a quarter of a mile in width. The depth at the front
was variable, from six to ten or fifteen feet. Its surface, both on
the top and at the front and sides, was bristling with ragged points
and edges of brownish-black a—-a. In this surface were innumerable
cracks, mouths, and ovens through which the red-hot matter shone
out. From many of these blue flames were flaring fitfully.
According to the writer’s observation, carefully concentrated on
this point, its mode of flow at this stage was as follows:
At the front, between the top and bottom, there was a slow,
forward, bulging motion of the intermediate layer, from four to ten
feet thick. As this progressed it produced fragmentation of the
thick, stiffened surface over it, pulling and breaking this away from
its contiguous parts at the more slowly moving top and bottom,
_and also breaking it up into an irregular mosaic with the changing
curvature of the surface of the front. At short intervals blocks of
the fragmented surface would be rotated into unbalanced positions,
when they would spall off from their own weight and drop to the
foot of the front, leaving for a moment bright, red-hot scars where
they had scaled away from the matrix within. Repeated examina-
tion of these scars showed continuous red-hot matter of very
viscous consistency, which cooled very quickly, tending only to
334 HARRY O. WOOD
bulge out a little without spurting or jetting. No cracks could be
made out on these freshly bared, red-hot surfaces. They exhibited
every appearance of a viscous continuum. Once they had crusted
over, however, the tendency to crack and bristle again became
noticeable. At this place the fragmentation and texture-forming
of these rough-surfaced blocks suggested very strongly the breaking
and cracking of candy pulled too long. There was no observable
gas action.
The whole effect here was one of creep, or overrunning, with
the plane of maximum rate of flow intermediate between the top
and the bottom. The action was that of a very viscous, fluid or
plastic, substance, flowing very slowly and exerting subsurface
traction upon a surface crust too stiff to draw or pull much.
It is thought that the matter was here cooled to a point where
it could still flow, or creep along, under blanketing, but once in
touch with the air it set so stiffly that any further strain frag-
mented it. Pieces artificially broken away from the bulging
surfaces cooled without further fragmentation, and without the
development of a—a-texture’ on those surfaces of the fragments
which were glowing when broken away. These exhibited the
rough fracture of cold basalt. The rough, pointed, and edged
texture of the natural a—a surface was seen in some cases to be due
to the drawing out of points and the shaping of rough edges as the
blocks were tilted and rotated away from each other, and from the
plastic matrix within, while the forward bulging movement was
taking place.
The mechanism of the formation of a—a has been considered a
very involved and complicated process, and the problems suggested
by it difficult of solution. The writer does not consider that the
observed action described above will serve to explain all cases and
details. But it does, he considers, indicate strongly that a-a
x A word concerning a—a-texture may not be out of place. As seen in Hawaii a-a
is not only block-lava, heaped flows of piled blocks and fragments of various shapes
and sizes, but each of these fragments is a separate unit, and its whole exterior surface,
in most instances, is characterized by an exceedingly rough aggregation of points,
edges, blades, spikes, knobs, etc., produced there in the process of the spalling and
transport of the blocks by the drawing, and possibly sometimes by spurting, of the
red-hot viscous basalt.
NOTES ON THE 1916 ERUPTION OF MAUNA LOA 335
generally results from fragmentation or granulation in the course
of flows of lava grown too stiff for further plastic flow under the
prevailing conditions—whether this results from rapid cooling due
to rapid escape of gas, to slow cooling, or to stiffening due to the
development of crystalline phases, throughout the mass, or in
lumps so as to form a sludge. The time is not ripe for discussion
and comparison of all these factors and their interrelationships.
But the action described above, with suitably conceived modi-
fications applicable in the region of more rapid streaming, and
where greater masses and higher temperatures are involved,
appears to the writer to be capable of explaining most of the vagaries
of surface texture and miniature surface forms exhibited by a—a in
Hawaii. Everything that the writer has seen in connection with
this outbreak, and all the reports that have come to his attention,
indicate that there was no excessive evolution of volcanic gases from
the molten lava on this occasion; and all indications are that the
temperature of the melt has not been excessively high. This points
to a relatively cool and viscous fluid. And so far this supports the
view of its action sketched here. It is, of course, not unlikely that
still other mechanisms are involved in the forming of a—a, and
emphatically there is no disposition to question any which rest
upon observation or sound rationale. However, to the writer it
seems unnecessary to appeal to unobserved, recondite, special
mechanisms to explain the fragmentation and textural qualities
of a—a. This view is in accord with the tenor of a view expressed
to the writer by Dr. William T. Brigham, of the Bishop Museum
in Honolulu, that a—a is the slush ice, or floe ice, in a cooled and
freezing stream (the granulation by motion of a stiff, overcooled
fluid on the point of solidifying). This seems the best short
expression of the idea.
The writer saw the action just as the stream had slowed up,
almost undoubtedly on account of failing supply at the source of
this branch. Thus the failure of pressure from above and the
radiation of heat all along the course led to rapid increase of
sluggishness. At this stage of the flowing there was practically no
gas action at the front. Blue fumes were rising from the surface,
along with heat-disturbed air, but these were so thin that from a
336 HARRY O. WOOD
tree near the front the writer was able to look a long way up the
flow. The only hindrance to good seeing was the shimmer of the
air produced by the heat radiation. The surface showed many
oven-like openings, and a few small conelike forms were seen, but
these were of temporary nature, and not true cones. No explana-
tion of their formation occurred to the writer. One that was
watched was slowly destroyed as the forward motion progressed.
The falling blocks made a tinkling sound, and the forward
motion of the upper surface was accompanied by a low grinding
sound, but these noises were low and inaudible at a short distance.
The quiet character of the advance at this stage was very striking.
At intervals loud detonations were heard. These were ascribed
to the action of the hot lava on buried vegetation.’ The sounds
made by the crackling of the falling trees and bushes were the
loudest of the frequent noises, and the crackling produced by the
burning of green vegetation was the most continuous and con-
spicuous. At this stage of the flow its approach was so quiet that
it gave practically no warning at a distance of fifty yards. ‘Trees
were being felled by the flow, partly by burning through at the
stump, but in some instances by overturning as a result of the for-
ward motion of the flow.
There were smells of subliming sulphur, sulphur acids, and of
cinders and charcoal. None of these was strong enough to be very
annoying. Others reported the smell of coal gas. This was not
noted by the writer, but was noticed by a large number of people,
and the fact must therefore be accepted. This was in the wooded
region, and here these carbon-gas odors could be ascribed to the
action of the lava on vegetation. However, along the Kahuku
branch of the flow such carbon-gas odors were plainly noticed by
many at points well above the wooded: region, where vegetation
was so scanty as to be negligible. And some have reported noticing
these odors near the lower end of the source in a barren region
where there is strictly no vegetation. It seems, therefore, that
carbon gases almost unquestionably were emitted from the lava
of this eruption.
[To be continued|
JourNnAL oF GrEoLocy, VoL. XXV, No. 4 PLATE}
Honoipy SETS
EO fool Pr soo 98°40 Longitude West from Greenwich
OY 4
Mohukone P
Failua Bp
K
I)
Hoopuloa PRZZZF, , . Ap
Honomalino Zs ye vg 7
a Or 2 HAWAII
Kapua >” YAW. Compiled from Govt Survey Maps.
by
Baldwin & Alexander
Civil Engineers
MILES
10
1907
Map of Hawaii, showing diagrammatically in red the flow of 1916, and the upper
outbreak source, with older flows in black. On this map the upper portion of the
flow of 1907 is not indicated precisely. This passed down the mountain on the west
side of Puu o Keokeo.
ee
lente
car Ne
oa : :
Bray
inst
JournaL oF GeoLocy, VoL. XXV, No. 4 PLATE II
Sik Y
Views a, b, c, and d snow successive stages in the development of the fume cloud [a at 7:45, 0 at 8:00,
c at 8:30, and d at 9:05 A.M.] of the earlier outbreak on May 10, 1916, as seen at a distance of about 25
miles, in a direction a little south of west, from the Hawaiian Volcano Observatory.
iat
Asif
ty Wei
JourNnaAL oF Grorocy, Vor, XXV, No. 4 PuateE III
a b
C d
a, showing a later stage, at 10:00 A.M., in the spread of the fume cloud of May 19, 1916.
b, showing the fume cloud above the head of flow at 8:30 A.M., May 22, 1916, as seen at a distance of
30-35 miles, ina direction S. 66° W. from the Hawaiian Volcano Observatory.
c, showing the brightly illuminated arch, or ‘‘primary glow,’’ above the source, and the course of the
Honomalino stream, as seen just before midnight, May 22, 1916, at a distance of about to miles in a direction
about N.N.W. from near Kahuku. This illuminated band or arch was about 7 miles in length and about
3 mile in height. Besides it, a bright, diffused red glow covered the entire clouded sky
by the view.
an effect not shown
d, a view from the Hawaiian Volcano Observatory exposed from 1:15 to 1:45 A.M., May 25, showing the
illurainated fume cloud above Kilauea (lower left) at a distance of 23 miles from the camera, of about 1,500
feet spread, and the illumination above the Kahuku stream (upper right), partly hidden by clouds at the
south, distant 30-35 miles, with a spread of about 5 miles and a height of a little less than 13 miles at
maximum,
; ce ny
mete?
xin
AGE AND STRATIGRAPHIC RELATIONS OF THE
OLENTANGY SHALE OF CENTRAL OHIO, WITH
REMARKS ON THE PROUT LIMESTONE AND SO-
CALLED OLENTANGY SHALES OF NORTHERN
OHIO
AMADEUS W. GRABAU
Columbia University
The name “Olentangy shale” was given by N. H. Winchell’
in 1874 to the light-gray soapy shales which underlie the Ohio
shale and are exposed on the Olentangy River and its tributaries in
central Ohio. Winchell regarded this shale as belonging with
the Huron shale which overlies it, but since his day this deposit
has usually been classed with the Middle Devonian,? and has been
made in a general way the equivalent of the Hamilton formation of
New York. The reason for such a grouping seems to have been the
fact that in northern Ohio a shale and limestone series lies between
the Huron shale and the Delaware limestone and so holds essentially
the position occupied by the Olentangy shale in central Ohio. I
have elsewhere? proposed to call this series in northern Ohio the
Prout series, from its exposures near the station of that name. In
the summer of 1914 I made a detailed study of the several outcrops
of this formation in the region about Sandusky, collecting at Plum
Creek and paying special attention to the contact of the Prout and
Huron formations shown in the exposures at Slate Cut on the
Lake Shore Railroad, halfway between Sandusky and Huron, and
in the “Deep Cut,” about a mile northeast of Prout Station. A
summary of my observations was included in the Report on the
Devonic Formations of Michigan submitted toward the end of that
1 Geol. Surv. Ohio, II, Pt. 1, pp. 287-80.
2 The author prefers the term ‘‘Devonic” to “‘Devonian,” but has changed it in
conformity with the usage of the Journal.
3 “Olentangy Shale of Central Ohio and Its Stratigraphic Significance” (Abstract),
Bull. G.S.A., XXVI (1915), 112.
337
338 AMADEUS W. GRABAU
year, but unfortunately not yet published. I quote from the
manuscript:
The upper 44 to 5 inches of the Prout has a peculiar character in that it is
full of pyrites, is irregularly bedded, and contains much glauconite. Black
shale specks and fish teeth are found in the upper half-inch. This upper part
of the limestone suggests a weathered and reworked portion very different
from the lower part, which is also dolomitized. Some doubtful limestone
pebbles have been found at the contact line in the base of the Huron shale,
but they are not sufficient in number to be of much value. Altogether, the
evidence is inconclusive, but it is not against the assumption of a disconformity
[between the Prout and the Huron]. ... .
The abrupt contact and the absence of intergrading are further
indications of a pronounced change in sedimentation with a long
time interval between the two formations. A comparison of the
fauna of the Prout with that of the Traverse group of Michigan,
gone into at some length in my report, shows the former to corre-
spond to the lower Traverse of Michigan, i.e., to the beds below
the Alpena limestone. I quote again from my manuscript report:
This means that the upper beds were never deposited or that they were
removed by erosion prior to the deposition of the black shale, for no one
would consider the black shale in any way contemporaneous with the upper
Traverse beds of Michigan. Thus an unquestioned time interval is indicated,
and since we find elsewhere the black shale disconformable upon the Traverse or
other Mid-Devonic beds, we need not hesitate to assume the same relation for
northern Ohio. ... . Compared with the sections in northwestern Ohio and
in Canada, the evidence becomes quite conclusive that between the Prout and
the Huron there is an unrecorded time interval.
Quite recently Dr. Stauffer’ has returned to a discussion of the
correlation of the Prout formation on the basis of its fossils, which
he listed in an earlier publication.2, He comes to the conclusion
that the Prout limestone represents the Encrinal limestone of
Eighteen Mile Creek,3 and the shales below it, the lower Hamilton
shales of western New York.’
™C. R. Stauffer, ‘‘The Relationships of the Olentangy Shale and Associated
Devonian Deposits of Northern Ohio,” Jour. Geol., XXIV, No. 5 (July-August, 1916),
pp. 476-87.
2 Geol. Surv. Ohio. Bull. No. 10, 4th Series, 1go09.
3[ have proposed the name Morse Creek limestone for this Encrinal of western
New York at the meeting of the Geological Society of America, December, 1914,
and in the report on The Devonic Formations of Michigan above referred to. It is an
older limestone than the Encrinal or Tichenor of central New York. See Bull.G.S.A.,
XXVI (1915), 113.
4 Now designated the Wanakah shales by me.
ar ae
OLENTANGY SHALE OF CENTRAL OHIO 339
That exact correlation with the Encrinal limestone of Lake
Erie is possible may perhaps be doubted, since the calcareous beds
increase in number westward. The Encrinal (Morse Creek) lime-
stone is the attenuated eastward extension of the great Alpena
limestone of Michigan, and the Prout limestone probably represents
one of the lower Traverse limestones of Michigan. Still, Stauffer
is undoubtedly correct when he makes the age of the Prout lime-
stone and associated shales lower Hamilton, and it is gratifying to
me to feel myself in substantial agreement with one who has made
such prolonged studies of these formations and faunas.
When it comes to the Olentangy shale of central Ohio, however,
Stauffer and I are in cordial disagreement. He makes it the equiva-
lent ‘of the Prout limestone and shales of the north and so of
Hamilton age, while I regard it as a part of the Huron shale series,
and referable to the Upper Devonian.
Although I had held this view for many years, it was not until
the summer of 1914 that I was enabled thoroughly to test my con-
clusions in the field. At that time I examined all the important
exposures of the formation in Delaware County, beginning with
Winchell’s type locality, on the Olentangy River. A new section
opened here for commercial purposes made a careful study possible.
The actual: contact between the Olentangy and Huron is sharp,
but perfectly even and uniform. In the upper portion of the gray
Olentangy are several thin bands of black or chocolate-colored
shale of the type of the Huron.
The bedding of the Olentangy shale is chiefly brought out by the
occurrence of thin bands of dark shale, and by more or less con-
tinuous layers of flat concretions. These are calcareous, up to
2 feet long by 1 foot thick, but mostly smaller. They abound in
iron pyrites, as does also the Huron shale overlying. In some
sections, as in the Deep Run and Lewis Center and Bartholomew
runs, the lower part of the Olentangy shale contains thin bands
of impure limestone. In one of these I found fish scales. In all
the sections, however, are found the thin bands of black shale in
the upper part of the gray, thus indicating a transition of the one
formation into the other. At the contact with the first great mass
of Huron shale there are sometimes found indications of a slight
drying of the surface of the Olentangy, with the formation of cakes
340 AMADEUS W. GRABAU
or scales of dry, gray mud, which were then incorporated in the
black mud. This is just what we should expect if the deposition of
the gray muds had come to an end and sedimentation were renewed
by the influx of the black mud from another source. Essentially,
however, deposition here was continuous, and after the commence-
ment of the sedimentation of the black Huron mud, there was a
temporary recurrence of the gray sedimentation, so that we see
today a 10-foot bed with all the characters of the typical Olentangy
lying above a considerable thickness of black Huron shale. In
both the upper and the lower part of this interbedded mass of
Olentangy occur thin bands of black shale, as they do in the typical
Olentangy lower down.
The basal contact of the Olentangy and Delaware is not shown
in any section which I visited, but the Olentangy could be examined
to within a few feet of the contact. There is no interbedding of
the Delaware and the Olentangy; the change in material is abso-
lute. The concretionary limestones of the Olentangy are very
different in character from the calcarenytes of the Delaware
limestone. The concretions appear to be of the subsequent type
found in the gray Cashaqua shales of western New York, to which
the Olentangy shales bear the closest resemblance. Like them,
they are unfossiliferous, though fossils are found in some parts of
the Cashaqua. The barren nature of both of these shales is in
striking contrast with the highly fossiliferous character of the
Hamilton shales of western New York, Canada, Michigan and even
northern Ohio. A few fragmentary fossils have been found in the
calcareous beds, but these might easily be residual specimens
weathered from the underlying limestones and incorporated in the
new sediment. Such undoubtedly is the origin of the lenticular
bed of crinoid fragments found in the type section, which does not
exceed 5 inches in thickness. This is apparently a reworked mass
of crinoidal fragments dissociated by the weathering of an older
crinoidal limestone.
The relationships here presented admit of only one conclusion,
namely, that the Olentangy shale is a part of the Upper Devonian,
representing a special type of sedimentation, such as characterized
the early Upper Devonian sediments of western New York. Sedli-
OLENTANGY SHALE OF CENTRAL OHIO 341
mentation was continuous from Olentangy into Huron time, but the
Huron type alone is represented in northern Ohio, where by overlap
it rests upon the eroded surface of the Prout limestone. The latter
is absent in central Ohio, where either it was never deposited, or,
what is more likely, it was removed by pre-Huron erosion. This
erosion extended down to the Delaware limestone, though it is not
impossible that a part of the lower Prout series is represented in the
central area by the Delaware limestone itself. If the name “ Prout’’
is to be restricted to the limestone member of the northern series,
then the shale below it must receive another name. It certainly is
not Olentangy, which name belongs to the earliest Upper Devonian
formation of central Ohio. -In my report on The Devonic Formations
of Michigan I have proposed the name “Arkona beds” for the shales
lying below the Encrinal limestone of the Thedford, Ontario, region.
If, as Stauffer holds, the shales below the Prout limestone are the
equivalent of these Ontario shales, which he calls Olentangy, then
the name “Arkona” may also apply to them. It may be wiser, how-
ever, to refer to them as the Plum Creek shales, since the distance
between Arkona and Plum Creek is too great to permit of positive
identification. True, Dr. Shimer and myself correlated the
Encrinal limestone of Thedford with that of western New York, on
the basis of faunal characters, and this correlation may be correct.
At the same time, we now know that the Encrinal of western New
York (Morse Creek) and that of central New York (Tichenor) are
not the same beds. I have also shown" that the faunas of the
shales below the Morse Creek in Eighteen Mile Creek occur in the
shales above this limestone 60 miles to the east, where they are not
found below that limestone. I have also shown that this typical
Hamilton fauna is absent from the beds above the Morse Creek
limestone at Eighteen Mile Creek, there being thus a complete
inversion of faunas. On purely faunal grounds the shales below
the Morse Creek at Eighteen Mile Creek would be correlated with
the shales above that bed at Moscow and elsewhere in the Genesee
Valley. The explanation of this and the relation of the western
New York Hamilton faunas to the Thedford and Michigan Traverse
t“The Faunas of the Hamilton Group of Eighteen Mile Creek and Vicinity in
Western New York,” 16th Annual Report, N.Y. State Museum, 1898, p. 330.
342 AMADEUS W. GRABAU
faunas is fully set forth in the unpublished report referred to.
There, too, it is shown that the faunas of the Traverse group on
opposite sides of the state of Michigan differ materially, while
identification of equivalent limestones and shales between the two
sections is impossible. All of these facts would lend some force to
the suggestion that precise correlation of the Prout limestone and
Plum Creek shales with the Encrinal limestone and Arkona shales
of the Thedford region should not be too rigidly insisted upon.
Nevertheless, we may with Stauffer lay much stress on the presence
of the Bactrites layer at about 25 feet below the Encrinal at Arkona
and a similar distance below the Prout limestone at Plum Creek,
S.OHIO CENTRAL OHIO N. OHIO S.MICHI GAN
== Huron shale
Fic. 1.—A generalized north-south section through Ohio and southern Michigan,
showing the relations of the Olentangy shale and the Prout formation.
containing at both places pyritized Bactrites arkonensis and Torno-
ceras uniangulare, besides Nucula triqueter and Leda rostellaria.
Then, too, as Stauffer has shown, the faunas of the Prout limestones
of this and of the Encrinal of Ontario are very similar, the latter
containing over 75 per cent of the species found in the former. On
the whole, therefore, Stauffer’s position seems well taken, and we
may accept his correlation of the Prout limestone with the “En-
crinal’’ of Thedford and perhaps with the Encrinal (Morse Creek)
of western New York.
We cannot, however, use the name Olentangy for the shales
below these horizons, and therefore the Canadian term ‘‘Arkona
shales” is preferable. This name may be then applied likewise
to the shales of Plum Creek. The comparative study of the
brachiopods of these various shales, now in process, will throw
further light on the provincial relationships of these formations.
OLENTANGY SHALE OF CENTRAL OHIO 343
Let us return once more to the typical Olentangy shale of central
Ohio, which we have seen is of Upper Devonian age. It rests
disconformably upon the Delaware lmestone, which represents
some of the lower Traverse beds of the Michigan region. There
is thus a great hiatus between the Delaware limestone and the
Olentangy shale in central Ohio, cutting out the greater part of the
Traverse group. This hiatus increases southward, so that in
Pickaway County the Olentangy shale lies in places upon the lower
Columbus and elsewhere upon the Monroan, and in all cases it is
succeeded by the black Huron shale. At Bainbridge, in Ross
County, it even rests upon the Niagaran. The Olentangy is still
represented at Vanceburg, Kentucky, on the Ohio River and near
Fox Springs, Fleming County, Kentucky, according to W. C.
Morse. A generalized north-south section through the region
named brings out the magnitude of the post-Traverse erosion, and
also shows that the Olentangy shale is of the nature of a lentil,
disappearing to the north and to the south. The source of the
Olentangy was probably local and circumscribed, representing
perhaps an accumulation in Upper Devonian time of a residual
soil produced from the weathering of the underlying rocks. It may
possibly be an extension of some of the eastern gray shales of the
Upper Devonian, such as the Cashaqua. The Black Huron shale
I hold to be a deposit of carbonaceous mud washed into the shallow
Upper Devonian sea by the rivers coming from the Devonian
peneplanes to the south and representing probably our best case
of an estuarine deposit in the American Paleozoic. The details
of this and the relation of the Huron to the Chattanooga shale,
which latter I consider mostly a terrestrial residual soil, reworked
in Mississippian time by the encroaching sea, are set forth at
length in the monograph on The Devonic Formations of Michigan
to which reference has several times been made.
_ THE HISTORY OF DEVILS LAKE, WISCONSIN
ARTHUR C. TROWBRIDGE
University of Iowa
The vicinity of Devils Lake, Wisconsin, has long been used as a
field of instruction in geology in the Middle West. The pre-
Cambrian igneous, sedimentary, and metamorphic rocks, the
Paleozoic history, the economic products, the general results of
glaciation, the origin and history of the lake, have all been reported
in a general way.’ It has been the privilege of the writer to con-
duct several courses in the district, to work over the data collected
and reported by previous investigators, and to work out additional
points not completed by them. He finds that the general history
of the district may be incorporated in an account of the history of
the lake and its basin.
The purpose of this paper is to bring together all the events
in the history of the lake basin, and to include some heretofore
unpublished conclusions as to the history of the lake, its past and
present sources of supply, and outlets once established but now
abandoned. This involves repetition of some facts already pub-
lished, and leads beyond the present boundaries of the lake basin.
LOCATION AND DESCRIPTION
Devils Lake is an almost rectangular body of fresh water 1535
miles long in a north-south direction and § mile broad, situated in
Sauk County, Wisconsin, 3 miles south of Baraboo, 40 miles north
by northwest of Madison, 80 miles southeast of LaCrosse, and 100
miles west by northwest of Milwaukee.
The district in which the lake lies is one of extraordinary high
relief for the central Mississippi Basin (about 800 feet) and of con-
siderable irregularity. The topography is dominated by two
«R.D. Salisbury and W. W. Atwood, Bull. No. 5, Wis. Geol. and Nat. Hist. Surv., ,
pp. 51-55; Samuel Weidman, Bull. No. 13, ibid., pp. 109-14; Lawrence Martin, Bull.
No. 36, ibid., pp. 177-78.
344
THE HISTORY OF DEVILS LAKE, WISCONSIN 345
ridges or ‘‘ranges,” known respectively as the North Range and
the South Range. These are the outcropping edges of the hard
Baraboo quartzite formation which forms here an asymmetric
syncline, the north limb of which is nearly vertical and forms the
North Range and the south limb of which has an average dip of
18° and forms the broader and higher South Range. The ranges
are 24 miles long in a general east-west direction, and they converge
at either end. The South Range is distinctly flat at and near
the top, the plain summit areas lying between 1,400 and 1,480 feet
A.T. There are also smaller areas of flattish land at 1,200 feet,
400 feet below Sauk Point, which is the highest point in the district,
and 4oo feet above main drainage lines. The summit of the
North Range is also flattish, its evelation being about 1,200 feet.
The district is drained by the Wisconsin and Baraboo rivers and
their tributaries. The Wisconsin River flows from the Dells, 17
miles north of Baraboo, in a broad curve past the east end of the
ranges at Portage to Prairie du Sac, where it may be said to leave
the district. Baraboo River enters the district at Ableman through
the North Range, in a gap known as the Upper Narrows, flows in a
general easterly direction between the ranges past Baraboo, cuts
back through the North Range at the Lower or Baraboo Narrows,
and joins the Wisconsin River near Portage.
Devils Lake occupies the northern portion of the only gap there
is through the South Range. This gap has the form of a broad
open curve with a north-south course in its northern portion, an
east-west course in the middle of the range, and a northwest-
southeast course at the south edge of the range. About one-fourth
of the length of the gap is occupied by the lake. The surface of the
lake is at 960 feet A.T. which is 160 feet above the Baraboo River
to the north, 200 feet above the Wisconsin River to the south, and
about 500 feet lower than the flat top of the range east, west, and
southeast of it (Plate I).
STRATIGRAPHY
The rocks of the district about Devils Lake include the Huro-
nian, Cambrian, and Ordovician, each system being represented by
more than one formation.
346 ARTHUR C. TROWBRIDGE
The oldest rocks of the district are granite, diorite, and rhyolite,
which lie around the borders of the quartzite syncline and seem to
form the basement upon which the quartzite was deposited. Next
above the igneous rocks lies the Baraboo quartzite, which in turn
is overlain conformably by the Seeley slate and the Freedom forma-
tion. The Baraboo, Seeley, and Freedom formations are all in-
volved in the folded structure of the district, although the Seeley
and Freedom beds do not outcrop. The total thickness of the
Proterozoic formations is about 6,000 feet, the quartzite alone
measuring in the North and South ranges about 5,000 feet.
There is a great unconformity between the Proterozoic and
Paleozoic groups of rock. After the Proterozoic formations were
deposited and folded, the region was eroded in one or more cycles
until the surface in this district had a relief of at least 1,300 feet, and
upon this rugged surface the Cambrian rocks were deposited.
The Cambrian system includes the Potsdam sandstone, between
600 and 700 feet thick where thickest, the Mendota limestone, 3-11
feet thick in the Devils Lake district, and the Madison sandstone,
80-90 feet thick. This is a conformable series and the strata are
essentially horizontal. Away from the ranges the top of the
Madison sandstone is about 1,020 A.T., but the standsone laps up
on the ranges to altitudes of 1,200 feet or higher. |
Only the Prairie du Chien and St. Peter formations of the Ordo-
vician system now appear in the vicinity of Devils Lake, although
it is nearly certain that younger formations were deposited here
originally and have been eroded away.
The Prairie du Chien formation overlies the Madison sandstone
conformably, but outcrops in only a few localities’ within the
boundaries of the Devils Lake district. It is hard, cherty dolomite.
Its average thickness in the Mississippi Valley is about 200 feet,
although in the immediate vicinity of Devils Lake its greatest
thickness is 20 feet.
An unconformity is known to exist between the St. Peter and
Prairie du Chien formations in Iowa, Illinois, Wisconsin, and
Minnesota, and is well represented in the Devils Lake district.
If the total thickness of 200 feet of Prairie du Chien dolomite was
THE HISTORY OF DEVILS LAKE, WISCONSIN 347
deposited here, all but 20 feet was eroded away a mile southeast
of the Baraboo Narrows, before the deposition of the St. Peter
sandstone. At Gibraltar Rock, 8 miles southeast of Devils Lake,
there is a thickness of 73 feet of Prairie du Chien dolomite between
the St. Peter and Madison sandstones on the southeast slope of
the hill and none at all on the southwest slope, the St. Peter sand-
stone lying directly on the Madison sandstone on the southwest side.
It is believed that the Prairie du Chien dolomite was entirely
removed in other places also, before the St. Peter sandstone was
deposited, as on the hill a mile south of the Pewits Nest (see
Plate II). The St. Peter formation is a massive, medium-grained,
quartz formation so similar to the Madison that the two cannot be
separated on lithologic grounds. The St. Peter sandstone is
thought by some geologists to be of eolian origin, but most of it at
least seems to be marine. In thickness the St. Peter varies from a
few feet to more than 200 feet within the district. The variation
is due to the erosion surface on which it lies, and to post-St. Peter
erosion of its surface.
Platteville limestone, Galena dolomite, and Maquoketa shale
are all found at Blue Mounds, 26 miles to the south of Devils Lake,
and its seems certain that these formations once covered the Devils
Lake district. If so, they were eroded away in some late Paleozoic,
Mesozoic, or Cenozoic erosion cycle, leaving no trace of their
previous existence.
The previous existence of rocks of Silurian age in the district
is proved by the finding of Niagaran fossils in a gravel deposit on
the summit of the quartzite range east of Devils Lake.'
Aside from the late Ordovician and mid-Silurian rocks which
once undoubtedly covered the region around Devils Lake, it is
entirely possible, though not proved, that formations of Devonian
and Carboniferous age were deposited also and were subsequently
eroded away. Certain it is that thick deposits of rock were laid
over the St. Peter sandstone.
Glacial drift and lacustrine deposits of, late Pleistocene age lie
unconformably on all the older rocks of the district (Plate II).
TR. D. Salisbury, Jour. Geol., III, 655-67.
348 ARTHUR C. TROWBRIDGE
THE EARLIEST RECORD
The history of the depression in which Devils Lake lies goes
back to pre-Cambrian times. Cambrian sandstone lies in the gap,
(1) at the north end of the east bluff one mile south of the northern-
most outcrop of quartzite in the South Range, 240 feet above lake-
level and 270 feet below the summit plain; (2) near the foot of the
west bluff, one-fourth mile north of the north end of the lake, 60
feet above lake-level and 450 feet below the tops of the bluffs;
(3) at the southwest corner of the lake at ‘“ Messenger’s End,”
where it extends from a few feet above lake-level to an altitude of
1,200 feet at the divide between Devils Lake and Skillett’s Creek;
and (4) forming a hill or bluff on the north wall of the gap 235 miles
east of the lake. At the last-designated point the sandstone
extends to goo feet A.T., which is below lake-level. The presence
of Cambrian sandstone at levels near that of the present lake
surface, near the north end, near the southeast end, and near the
center of the gap shows clearly that there was a depression or that
there were depressions here before the advance of the Cambrian
sea. The facts might be interpreted in one of two ways:
1. The simplest explanation lies in the assumption that after
the pre-Cambrian formations were folded, the surface was eroded
- and reduced to a peneplain on which a river meandered, that this
plain was uplifted relative to the sea, and that the district had
reached late youth or early maturity in a second cycle of erosion
before submergence by the Cambrian sea. Under this interpre-
tation the original gap is a pre-Cambrian intrenched meander.
The ancient peneplain may be represented by the present summit
plain of the district, or it may have been entirely destroyed in sub-
sequent periods of erosion.
2. The conditions, however, might be almost equally well
explained by assuming that the district was in maturity either of
its first or of some later cycle of erosion when the Cambrian sea
advanced, that at that time a deep valley existed in the range with
its head in the valley of Messenger’s Creek and discharging east-
ward and southeastward, and that another stream headed some-
where north of Messenger’s Creek and flowed northward into the
Baraboo valley. This would not necessitate more than one cycle of
THE HISTORY OF DEVILS LAKE, WISCONSIN 349
erosion in pre-Cambrian times nor would it preclude more than
one cycle.
After the long series of events during the Paleozoic and subse-
quent history of the district, it does not seem possible to determine
which of the foregoing interpretations is correct. However, some
light is thrown upon the matter by consideration of the pre-
Cambrian history of Wisconsin, and the principles of stream adjust-
ment. It has been demonstrated by Weidman’ and Martin’ that
the surface of Wisconsin was degraded to a peneplain in pre-
Cambrian times, that this plain slopes south and is buried beneath
Cambrian sediments at an altitude of about 300 feet in the latitude
of Devils Lake, and that the quartzite ranges stood as erosion
remnants on this plain. Well-records in the Baraboo valley show
that the pre-Cambrian surface within the inclosure made by the
ranges is as low at least in some places as 340 feet A.T. A stream
must therefore have had entrance to the inclosure and a means
of escape from it. A broad, continuous, stream-made gap, filled
with Cambrian sediments, cuts through the North Range north-
west of Baraboo, and is probably the line of entrance or exit of a
large pre-Cambrian stream. The other gap, either entrance or
exit, is not known unless it be the Devils Lake gap. Neither the
Lower nor the Upper Narrows seems to be large enough and they
have not yet been proved to be pre-Cambrian in age. It is improb-
able that there is a buried gap, undiscovered, either in the North
or the South Range, which might have conducted the river in or
out of the area between the ranges. It is unlikely also that two
streams would adjust themselves as postulated in the second case
above. A stream working headward into a hard, high ridge with
a steep slope would hardly develop a course other than one more or
less nearly straight and nearly normal to the trend of the ridge.
But the stream flowing southeastward from Messenger’s Creek
must have had a course which was distinctly curved, had supple-
mentary angles of 30° and 150° with the trend of the ridge, and
was oblique to the dip of the rocks. On the other hand, the
t Samuel Weidman, Bull. 16, Wis. Geol. and Nat. Hist. Surv., pp. 385-95, 592-600;
Jour. Geol., X1, 289-313.
2 Lawrence Martin, Bull. 36, Wis. Geol. and Nat. Hist. Surv., pp. 347-73.
350 ARTHUR C. TROWBRIDGE
freshness and apparent youth of Devils Lake gap, and its size,
harmonious with the dimensions of the Upper and Lower Narrows,
suggest that these three gaps were cut at the same time and during
some post-Silurian period. And yet Devils Lake gap may have
carried a large river in pre-Cambrian times and the gap may have
been reoccupied at a later time when the two Baraboo narrows
were cut.
Whether the pre-Cambrian Devils Lake gap was cut by a single
intrenched meandering stream, or by two streams with a col between
them, the depression must have been a deep one. The rim of the
canyon is today represented by the 1,470-foot summit plain, and the
pre-Cambrian rim may have been higher than this. Cambrian
sandstone in the gap is known as low as goo feet A.T. The depres-
sion must therefore have been at least 570 feet deep in pre-Cambrian
times. Certainly if the gap was cut by the same stream which
reduced the inclosure between the ranges to 340 feet A.T., and
probably if it was cut by two streams tributary to main drainage,
the bottom of the gap at the end of pre-Cambrian times was
not much higher than 340 feet A.T. The pre-Cambrian gap was
then probably not far from 1,100 feet deep. The whole district
at this time is known to have had a relief of 1,200 or 1,300 feet.
PALEOZOIC HISTORY _
The Paleozoic, marine, sedimentary rocks of the district
record the second important step in the known history of the
district.
Upper Cambrian sandstone is found on the ranges to an altitude
of 1,200 feet A.T. and probably exists at slightly higher levels.
On Wood’s Quarry Hill, 3 miles northwest of the lake, rounded
pebbles of quartzite are found in the uppermost beds of the Madison
formation where it is overlain conformably by Ordovician dolomite,
giving evidence that quartzite was exposed on near-by land up
until the very end of the Cambrian period. In the light of these
facts, it is believed that the South Range was an island in the
Cambrian sea, and that the Devils Lake gap was not filled to a
higher level than 1,020 feet, although the sea seems to have reached
at least to 1,200 feet.
THE HISTORY OF DEVILS LAKE, WISCONSIN 351
But deposition did not end with the Cambrian period. The
Prairie du Chien and St. Peter formations, or their time equivalents,
must have been deposited in the gap. It is clear that the sediments
of the Prairie du Chien stage did not fill the gap, for the base of
this formation has an-elevation of 1,020 feet in the vicinity, and it
would require a thickness of 500 feet to have filled the gap. As the
formation is nowhere known to be so thick, it is probable that a sag
existed at the site of the present gap, when the Prairie du Chien sea
had withdrawn, and it is possible that the gap was again occupied
by running water and partly re-excavated before the deposition
of the St. Peter sandstone. As the St. Peter sandstone is not ,
found much above 1,300 feet in the district, it seems clear that the
gap was not entirely filled with this deposit.
Although no traces of the Platteville, Galena, and Maquoketa
formations are found in Devils Lake gap or in its immediate vicinity,
it seems clear that these formations, or their time equivalents, were
deposited in the gap, filled it, and buried it, for if the dip of these
formations be projected northward from their existing altitude at
Blue Mounds and other points to the south, the bottom of the
Maquoketa formation would lie 200 or 300 feet above the summit
plain on the South Range. Gravels containing Niagaran fossils
are found to the very top of the range on the flat summit plain east
of the lake, showing that this formation added its thickness to the
sediments which buried the filled gap and the ranges.
If still younger Paleozoic formations were deposited over the
district and over the filled gap, they have left no record. So far as
the records go, the Paleozoic sea retreated finally at some date after
the Niagaran epoch. These seas probably left the district essen-
tially flat, with the rough surface of the quartzite buried beneath
thick Paleozoic sediments. At this time there was no Devils Lake
gap as a topographic feature, but there was a previously existing
gap, filled with sediments and buried by formations which also
covered the ranges deeply.
POST-PALEOZOIC—-PRE-GLACIAL HISTORY
The record of the history of the gap following the final with-
drawal of the Paleozoic seas and antedating the advance of the
352 ARTHUR C. TROWBRIDGE
Wisconsin glacier is to be read only from the study of the topography
and surficial deposits of this and surrounding districts.
The flat summit areas at various places on the South Range at
altitudes between 1,400 and 1,480 feet, lying across the beveled
edges of the quartzite beds, can be interpreted only as an ancient
plain of degradation now almost destroyed by streams working in a
later erosion cycle. The peneplain might be considered to be of
pre-Cambrian age but for the fact that there are stream gravels on
its surface which are composed of Paleozoic rocks, and contain
fragments in which are imbedded Niagaran fossils. Associated
with the gravel there are numbers of potholes which are not filled
by Paleozoic sediments and which probably are of post-Niagaran
age. The idea that the flat is a remnant of the old pre-Cambrian
plain buried by Paleozoic formations, resurrected by later erosion,
and becoming again the site of deposition as a part of the later pene-
plain, is perhaps tenable, although hardly probable, for the extension
of this plain has now been traced west into Iowa and south into
Illinois, in both of which states it cuts across the beveled edges
of the Potsdam, Prairie du Chien, St. Peter, Platteville, Galena,
and Niagara formations in order.
It is not to be understood that this erosion surface had reached
a final stage of degradation and was perfectly flat. It is made
clear by a study of the Devils Lake district, as of other districts
where the plain is known, that it had considerable relief. West
of Devils Lake the surface of the plain is 1,400 feet A.T., which seems
to be the altitude of the portion which was brought to grade. East
of the lake the gravels lie on a flattish surface at 1,470 feet, and
quartzite at Sauk Point reaches an altitude of about 1,600 feet.
Before this surface was dissected, it had a relief of 200 feet in the
Devils Lake district, and the gravel occupies a position between the
lowest and highest portions of the surface. The gravels probably
were not brought from a distance by a long and large stream at
grade, but were more likely to have been deposited by a stream
tributary to main drainage, the tributary having enough velocity
to carry gravel and to cut potholes. The gravels include no
material which could not have been derived from local formations.
On the other hand, the relief of the surface must have been much
THE HISTORY OF DEVILS LAKE, WISCONSIN 353
less than that of the present surface, and Devils Lake gap could
not have existed at the time.
The exact age of this erosion surface is still an open question.
Where the plain and the gravel associated with it are known outside
this district, they have been assigned by different writers to differ-
ent ages. Winchell? long ago correlated these gravels in Minne-
sota tentatively with the Cretaceous, and following his lead Bain,
Calvin,; Grant and Burchard,’ Hershey,’ and others have con-
sidered the plain to have been formed chiefly during the Cretaceous
period. This correlation is, however, somewhat doubtful, for the
reason that it has never been proved that the gravel on the plain
in Minnesota is of Cretaceous age. It seems to lie on the Cre-
taceous, a relationship which tends to show that the plain cuts
across the edges of the eroded Cretaceous rocks and is therefore
post-Cretaceous in age, just as the fact that it cuts across the
‘Paleozoic sediments proves that it is younger than those sediments.
Most likely the plain is of Tertiary age. The evidence of this has
been presented by Salisbury® in an article in which he suggests the
correlation of these patches of gravel with the Lafayette formation
farther south.
It is at least clear that the Devils Lake gap remained filled with
Paleozoic sediments while this plain was being formed, and that
the time involved was long.
There are strong suggestions of a second flattish erosion surface
with remnants at about 1,200 feet. Representatives of this
surface may be found (1) cutting across the beveled edges of the
vertical quartzite beds of the North Range at the Upper Narrows,
(2) at the Lower Narrows, (3) 2 miles northeast of Denzer, (4) on
the summit of Old Flat Top, 1 mile southeast of the Lower Nar-
rows, (5) on the top of Gibraltar Rock, 13 miles west of Okee, and
(6) forming saddles or low divides in the South Range, as on the
tN. H. Winchell, Geol. and Nat. Hist. Surv. Minn., I, 309-31, 353-50.
2H. F. Bain, Bull. U.S. Geol. Surv., No. 294, pp. 11-16.
3 Samuel Calvin, Jowa Geol. Surv., IV, 43.
4 Grant and Burchard, Lancaster Mineral Point Folio, U.S. Geol. Surv., p. 2.
5O. H. Hershey, Am. Geol., XX, 246-59.
6R.D. Salisbury, Jour. Geol., III, 655-67.
354 ARTHUR C. TROWBRIDGE
divides between the North Fork of Messenger’s Creek and Skillett’s
Creek, between the South Fork of Messenger’s Creek and an
unnamed south-flowing stream, between Pine Creek and Otter
Creek, etc. Because the remnants of this plain within the district
cut across the quartzite beds and lie on St. Peter sandstone, and
because most of the streams of the district find their sources on
the plain, the remnants are believed to represent parts of a plain
of erosion developed at this level, but now mostly dissected.
This interpretation is greatly strengthened by the finding of a
similar plain, bearing the same relation to the older plain, and
cutting across the edges of eroded formations, at many places out-
side the district under consideration, as, for instance, in the Rich-
land Center quadrangle, in the Sparta quadrangle, in the Lancaster
and Mineral Point quadrangles, in the eastern portions of the
Waukon and Elkader quadrangles in Wisconsin, in Jo Daviess
County, Illinois, in Allamakee, Clayton, and Dubuque counties in
Iowa, and at various places in southeastern Minnesota.’
If this plain is correctly interpreted, it records the following
steps in the history of the district in general and of Devils Lake
gap in particular. After the 1,400~-1,480 foot plain was developed,
the district was uplifted relative to the sea-level of that time to an
amount of approximately 200 feet, the streams were rejuvenated,
and reached grade again at levels 200 feet lower than the first
plain.
It was during this cycle of erosion that Devils Lake gap was
re-excavated and the Upper and Lower Narrows were formed, or
re-formed if they are of pre-Cambrian age. In the formation of
these post-Paleozoic gaps problems of stream adjustment are
involved. Martin? expresses disbelief in the two peneplains in this
part of the country, considers that all post-Niagaran and pre-Wis-
consin erosion took place in a single cycle, and believes that the
Devils Lake gap, and the two Baraboo gaps in the North Range,
tO. H. Hershey, Am. Geol., XX, 246-59; U. B. Hughes, Proc. Iowa Acad. Sci.,
XXIII, 125-32; W. D. Shipton, Master’s thesis, University of Iowa Library;
U.S. Grant and E. F. Burchard, Lancaster Mineral Point Folio, U.S. Geol. Surv.,
p. 2; A. C. Trowbridge and E. W. Shaw, Bull. No. 26, Ill. Geol. Surv., pp. 136-44;
A. C. Trowbridge, Bull. Geol. Soc. Am., XXVI, 76.
2 Lawrence Martin, Bull. No. 36, Wis. Geol. and Nat. Hist. Surv., pp. 63-70 and 177.
THE HISTORY OF DEVILS LAKE, WISCONSIN Bos
were made by the Wisconsin and Baraboo rivers developing their
courses on the surface of the Paleozoic rocks, cutting down through
these rocks and becoming superimposed on the ranges, and holding
their courses. In the case of Devils Lake gap, at least, this
hypothesis seems to involve too nice a coincidence. Whether the
pre-Cambrian gap was continuous or made up of two valleys with a
col between, Martin’s idea would mean that a crooked stream
starting on a surface 1,200 feet or more above final grade, cuts down
more than 300 feet, then develops a flat surface and deposits fine
gravel without ceasing to cut, is superimposed on quartzite in a
course exactly coinciding with a peculiar pre-existing filled and
buried crooked valley, and then cuts on downward for goo feet
without interruption.
It seems more likely to the writer that the explanation of the
reoccupation of the gap by a stream is to be found in the applica-.
tion of the principle of stream adjustment on non-resistant rocks
during a cycle of erosion which went nearly to completion. On the
1,400-1,480 foot surface, the streams flowed here on quartzite and
there on sandstone. With the uplift of the surface, these streams
began to intrench themselves and new tributaries were formed.
The larger streams reached grade after cutting 200 feet or so.
For most of the streams this downward cutting was through
quartzite. The stream which adjusted on the non-resistant
sandstone in the gap cut more rapidly than the others, obtaining
an advantage in this way; it became a pirate and gradually cap-
tured many of the other streams. That there were other streams
during this cycle and that they did intrench themselves before
being captured is proved by the fact that there are passes or cols
across the range at altitudes a little above 1,200 feet, as, for instance,
where the West Sauk road crosses the range between the heads of
Skillett’s Creek and Otter Creek.
At any rate, it was during this cycle of erosion that the Wis-
consin or its pre-Glacial ancestor came to flow southward and west-
ward over the present site of the Lower Narrows, up the present
course of the Baraboo valley and southward through the shallow
Devils Lake gap, and the Baraboo River entered the inclosure
between the quartzite ridges through the Upper Narrows, and
356 ARTHUR C. TROWBRIDGE
joined the Wisconsin at about its point of entrance to Devils Lake
gap. The Upper and Lower Narrows may have been formed for
the first time by superimposition during the formation of the 1,200-
foot plain, or possibly by adjustment of the streams on sandstone,
provided they had been made and filled with sandstone.
The age of this lower plain and therefore the date of re-
excavation of Devils Lake gap are no more proved than is the age of
the upper erosional surface. Those who hold that the upper plain
is Cretaceous assign a Tertiary age to the younger plain, and those
who believe the upper plain to be Tertiary in age naturally assume
that the lower plain was formed at the end of the Tertiary period
or early in the Pleistocene. There is some evidence in north-
eastern Iowa that this plain was intact when the oldest glacial
drift was deposited, but this also must be considered to be an open
question until more of the field data have been published. At
least it is clear that the present Devils Lake gap had its beginnings
either in late Tertiary or in early Pleistocene times, which for
present purposes is perhaps close enough.
Long before the monadnocks had been removed from this lower
plain—that is, before the second cycle of erosion had reached com-
pletion—there was another uplift of the land relative to the then
existing seas, and the streams were again rejuvenated. This
uplift was much greater than the previous one, for the valleys
cut during this third cycle of erosion are much deeper than any
which were cut during the second cycle. At this time, too, the
main part of the present Devils Lake gap was cut. The bottom
of the gap at the beginning of this erosion cycle could not have been
lower than 1,200 feet A.T. and the tops of the bluffs were no higher
than 1,470 feet; that is, the maximum depth of the valley up to
the beginning of the last erosion cycle preceding the deposition of the
glacial drift was 270 feet. The maximum depth of the valley at
the end of this cycle could be determined, if it were possible to get the
altitude of bedrock underlying the glacial deposits in the middle of
the gap, a bit of information which unfortunately is not available.
The deepest boring into the glacial material within the gap was
made by Gustaffson and Prader in 1914 at a point near the middle
of the gap at the north end of Devils Lake, only a few feet above
THE HISTORY OF DEVILS LAKE, WISCONSIN 357
lake-level. The altitude of the well-site is about 965 feet. This
well penetrates 283 feet of glacial material without striking rock;
therefore the bottom of the pre-Glacial gap must be somewhere
below 682 feet, and the pre-Glacial gap at this point must have
been at least 788 feet deep, at least 500 feet deeper than it was at the
end of the second erosion cycle, and at least 283 feet deeper than it
is today. The maximum depth of this gap might be known if the
altitude of the lowest sub-drift bedrock in the Baraboo valley
between the ranges or in the Wisconsin valley south of the South
Range could be obtained, assuming that a stream flowed from the
Baraboo valley through the gap to the Wisconsin valley to the
south. The lowest bedrock surface obtainable in the Baraboo
valley east of Baraboo is at 570 feet. Assuming that the bedrock
in the Devils Lake gap is as low, the pre-Glacial gap was at least
goo feet deep. It cannot be ascertained whether the pre-Cambrian
gap was deeper than this or not so deep, for it cannot be determined
whether there are Paleozoic sediments below the bottom of the
present gap, nor whether the tops of the ranges were higher then
than now. Ifit be true that the bottom of the pre-Cambrian gap lies
at or near 340 feet, the pre-Cambrian gap was probably some-
thing like 200 feet deeper than the pre-Glacial gap.
THE GLACIAL LAKE
So far as ascertained, no glacier prior to the Wisconsin glacier
affected the Devils Lake gap. There have been some suggestions of
pre-Wisconsin drift in the vicinity," but these evidences have
proved to be negative. It seems likely, however, that the Illinoian
glacier advanced almost to this district; but if it played any part in
the history of the lake or its basin, its effects are not now visible
within the district.
As has been brought out by Salisbury and Atwood,’ the Wiscon-
sin glacier formed Devils Lake and had a controlling influence in its
early history. As the ice moved into the district from the north-
east, it was divided by the ranges, one lobe advancing down the
t Samuel Weidman, Bull. No. 13, Wis. Geol. and Nat. Hist. Surv., pp. 99-102.
2R.D. Salisbury and W. W. Atwood, Bull. No. 5, Wis. Geol. and Nat. Hist. Surv.,
PP. 132-33.
358 ARTHUR C. TROWBRIDGE
old valley of the Wisconsin from the Lower Narrows to the horth
end of Devils Lake gap, where its edge became stationary and
deposited a terminal moraine, the other lobe coming in south of
the South Range and advancing up the valley of the pre-Glacial
Wisconsin to deposit a marginal ridge across the gap east of its
major bend (Plate IZ). This left the north-south portion of the
gap and a part of the east-west portion confined between the two
edges of the ice, and in the basin so made Devils Lake was formed.
Connecting the ends of the two lobes, the edge of the ice reached
its limits of advance in an irregular line crossing the South Range
from the north edge of Devils Lake eastward to Sauk Point, and
thence southwestward to the gap east of Kirkland (Plate II).
After its formation the lake had an interesting history during
the occupancy of the ice.
Sources of supply.—When the lake was first formed, as outlined
above, there were at least four separate sources of water supply:
(1) The edge of the glacier blocked either end of the lake basin.
There the ice melted and furnished water for the basin. Study
of the terminal moraine from the north end of the lake around by
Sauk Point and southwest to the gap east of Kirkland leads to the
conclusion that the water resulting from melting along this whole
stretch of ice front must have flowed into Devils Lake basin.
(2) The bottom of the basin was below ground-water surface, as
evidenced by the fact that it had been occupied by a permanent
stream up to the time when this stream was blocked by the ice,
and ground water was a source of supply. From the inception of
the lake until its bottom was built up above ground-water surface
by fluvio-glacial deposition, if this stage was ever reached, some of
the lake water may have come from under ground. (3) The lake
must have had inlets resulting from precipitation within the
borders of the lake basin. For instance, Messenger’s Creek with
its north and south forks must have flowed into the southwest
corner of the lake as it does today, and the stream which flows
west and north past the northeast corner of the lake, being blocked
to the north by the ice, must have contributed its waters to the
lake. (4) There was doubtless direct precipitation upon the
surface of the lake. Of these four sources of supply, the first men-
THE HISTORY OF DEVILS LAKE, WISCONSIN 359
tioned is conceived to be most important, and needs more complete
description.
From the extreme east end of the east loop of the terminal
moraine at the west foot of Sauk Point, the land slopes south from
the north limb of the moraine, north from the south limb, and west _
from the junction of the two limbs. The water formed by melting
at the edge of the ice must therefore have concentrated in the
depression between the two limbs of the moraine and must have
Jusaer
Point.
\
\\
Pat, SS
Tey ra \
Oy At A \ .
I NN ~ Glacier Ice
NS ey) <<
re A AT EL
Fic. 1.—Sketch map showing Devils Lake and its drainage basin during the
occupancy of the ice.
North
flowed westward into what has been called the Steinke Lake by
Salisbury and Atwood! (see Fig. 1). The same general conditions
existed in the minor loop north of the Steinke Lake, and the waters
from this loop must have mingled with those from Sauk Point in the
Steinke Lake.
The Steinke Lake was a body of water about 3 mile long east
and west and } mile broad north and south, held in by a low ridge of
quartzite on the west, the north limb of the ice edge on the north,
the westward slope of the land from Sauk Point on the east, and
the northward slope of the South Range and the south limb of the
tR. D. Salisbury and W. W. Atwood, Bull. No. 5, Wis. Geol. and Nat. Hist.
Surd., pp. 120, 133-34, Pl. XXXVII.
360 ARTHUR C. TROWBRIDGE
ice edge on the south. Short arms or bays projected east toward
Sauk Point and north into the narrow loop of the ice edge. As
discovered by Salisbury and Atwood, the lowest point in the basin
of this extinct lake during the occupancy of the ice, and hence the
head of the outlet of the lake, was at the extreme northwest corner
of the lake a few rods across the Merrimac Road west of the home
of Julius Steinke. The bottom of this outlet is at approximately
1,250 feet A.T., and as the outlet is broad and shallow, it may be
assumed that the level of the lake was little if any above 1,260.
The water from this lake flowed westward along the front of the
ice toward Devils Lake.
The materials now occupying the site of this extinct lake are
lacustrine silts, sands, and gravels, finely divided near the middle
and coarser around the borders, and coarser at the surface than in
the bottoms of deep cuts or borings. The maximum depth of the
original lake is not known, but a well on the north side of the flat
at the house next west of the schoolhouse, whose site is at 1,260
on lacustrine material, penetrates 202 feet of what appears to be
lacustrine material, without reaching rock. This indicates that
the original lake was at least 200 feet deep, and that the 200 feet
of debris deposited in it was sufficient to fill the lake basin by the
time of the retreat of the ice.
From the Steinke Lake the water drained westward into a small
pocket or basin having a flat bottom. In late years this little plain
has been known locally as the Peck flat. It is an area of perhaps
80 acres, bordering the terminal moraine on the north and sloping
gently and getting narrower to the south. On the west, east, and
south there are high hills of quartzite, but there is a break in the
rim of the basin at its southwest corner, through which drainage
is free to flow south and west to the north end of Devils Lake. The
width and depth of this valley, the hardness of the quartzite which
forms its walls and bottom, and the small size and intermittent
character of the stream which drains it, when compared with the
post-Glacial valleys of Skillett’s Creek and the Wisconsin River,
show that this outlet to Peck flat is pre-Glacial. Yet when all
available authentic well-records are considered, it is clear that
glacial waters could not have flowed at first through this outlet
THE HISTORY OF DEVILS LAKE, WISCONSIN
361
depression unobstructed. The accompanying table gives the
altitudes of well-sites and rock outcrops, depths to sandstone,
altitudes of bedrock, etc., for points located in Fig. 2.
TABLE OF WELLS AND OUTCROPS IN THE PECK BASIN
Ihitideon be Elevation of Denthite Elevation of
Wells and Outcrops : eee : aes Sandstone Ousecite Qusrizite
(Feet) (Feet) (Feet) (Feet) (Feet)
Johnson welll yi) ea: 1,280 113 1,167 ? ?
Reck drilled wellos 334); .) 13225 38 1,187 ? ig
Peck dug well.........- 1,220 26+ 1,194— ? ?
Trongdrillthole wa. ha v4 1,210 18 1,192 207-5 913—
Marquid iwellin )y.))22/. 1,215 6 1,209 T2 0 1,094 —
Steinke sandstone outcrop} 1,205 fo) 1,205 ? ?
Stemke/quartzite outcrop 1. b80, aise sates ns sates ° 1,180
Stemkeswellea ime) ase 1,218 6 1,212 155 1,063
From this table and from Fig. 3, it is made clear that the pre-
Cambrian surface of this section, as of all portions of the district,
was very irregular, and that the post-Paleozoic and pre-Glacial
surface sloped north and west from opposite sides of a divide
located somewhere near the south end of the present flat.
It is clear that the ice advancing from the north blocked drain-
age in that direction and melting, furnished water which joined
with the discharge from Steinke Lake to make a small lake in the
Peck basin. The waters of this lake rose rapidly to the level of
the divide at the south end of the basin and overflowed westward
into Devils Lake. Peck Lake, shallow from the first, was gradu-
ally filled until the lowest point on its bottom was as high as the
outlet, and the lake ceased to exist. The filling of the lake must
have been accomplished before the retreat of the glacier, for fluvio-
glacial material was deposited over the lacustro-glacial material.
The Peck dug well (see Fig. 2) penetrates 8 feet of coarse gravel
and below that 18 feet of sand. The gravel is fluvio-glacial and
the sand below lacustro-glacial. The top of the sand is at 1,212
feet A.T., and this is probably the approximate altitude of the
pre-Glacial divide.
From the foregoing it is apparent that all the water from the
edge of the glacier in its great complex loop east of Devils Lake
flowed into Devils Lake during the occupancy of the ice (see Fig. 1).
ARTHUR C. TROWBRIDGE
The Size of Glacial Devils Lake.—After description of the sources
of supply for the basin of Devils Lake during the occupancy of the
362
q
SS SSA)
SI
tite &&X
eee
SSS
ates
ndsto
ison Qvuartsite cr
ay 4
[7
BS
Se
SK
Z KLE
“ <>.
fo O° Oe is
2 090 0 OB
writers.
2 Poe on
> oe 5 :
© oP 82 20 SRS
00 2a oP Oo
pera \-2- bed Yo 5
iD ote
990 bPoY o.
: 2,
elon °
12.406 Dh,e [OM
eNpos EE Pe igo
7 eo A =50)
a.2- Ko lq g ko Ax
= pelo a7 7oa08
Ce eter Chee 2 Accin
'@:e-* uN ° re) Boo
oS 6 OP po? OF53
n > ty aq oF. 8!
=. : °
SN GS oye 099,
Bg.e& qe Bo°5
~e ge = 05°
e\. ge
g-\ f
P38 /P a
OPE o
gta? Oge
Lee fors pe
ESY 9,954 09
2 rod 2 0?
e 150 969 090
S659 9h oo
Fe 2940
PS
° 2
‘Hage Z
‘TSO
pee) Y)
O VA
YY
than
given, also the line of the
mpanying table are
Fic. 2.—A geologic and topographic map of the Peck basin. The locations of
and outcrops involved in the acco
wells
now after most of these sources have been cut off. Indeed, there
ice, it is evident that the glacial lake must have been larger
section 4A, shown in Fig. 3.
THE HISTORY OF DEVILS LAKE, WISCONSIN 363
is a question as to whether the basin of Devils Lake was large
enough to confine all this water. If it be assumed that the amount
of water supplied to the basin by ground water was balanced by
loss due to seepage into the débris at either end of the lake, and
that the precipitation on the surface of the lake and on the lakeward
Steinke Sandsfone Oufcrop
Steinke Quortzife Outcrop
Marquid = Well
Steinke Well
; $
= =
€ &
FS <
S =
Ss =
ZA Ne a, Ey POI
a ee Cs ETT | (ST
: ET SS) EE | | Ee a
ppsceett
VAR THE
~ ial
Huronian Quartrite &
isos ego
ns rs
kes
Fic. 3.—Section along line AA, Fig. 2, showing surficial and underground condi-
tions in the Peck basin.
slopes was balanced by evaporation, the glacial water is still
unaccounted for.
Careful measurement of the terminal moraine from the north-
west edge of the Devils Lake basin, through all its curves to the
crest of the Devils Nose southeast of the lake, shows that water
drained into Devils Lake from 11.6 miles of ice edge. To get some
conception of the amount of water which this front of ice con-
tributed, let us suppose that the ice edge tributary to the lake had
364 ARTHUR C. TROWBRIDGE
an average thickness of 100 feet (probably less than the actual
thickness) for the first mile back from its edge. To estimate
the total annual water supply from the ice, it is necessary to make
reasonable supposition as to the amount of annual melting measured
in a horizontal line normal to the front of the ice. The ice front
may be considered to have been essentially stationary, and the
amount of ice melted can be measured by the rate of glacial motion
in the marginal portions of the ice. Chamberlin and Salisbury*
estimate that the ice in the Greenland glacier moves something less
than a foot a week near its edge. Suppose that the glacier which
affected Devils Lake moved 6 inches a week or 26 feet a year. IH,
then, it be assumed that a volume of ice 11.6 miles long measured
along the edge of the glacier, roo feet thick measured vertically,
and 26 feet wide measured normally to the ice front, melted each
year and ran into Devils Lake, there would be an annual total of
about 14 billion cubic feet. The decreased volume due to change
from ice to water may be neglected in a computation where there
are so many assumptions and where all figures have been reduced
to a minimum for safety.
To estimate the capacity of the basin of Devils Lake during the
occupancy of the ice, it is necessary to have its length, width, and
depth. Measured from moraine to moraine around the curve
of the gap, the basin is almost exactly 2 miles long. The average
width from end to end and from lake level to lowest point in the
rim of the basin is approximately 3 mile. The depth of the glacial
lake may be found by subtracting the altitude of the present lake
bottom from the altitude of the lowest point in the rim of the basin,
_ neglecting the glacial débris below the bottom of the lake, which
would have displaced the water as it was deposited. Computed
in this way, the maximum depth of the glacial lake was 270 feet.
Multiplying the depth, width, and length, the capacity of the
glacial basin was about 7% billion cubic feet.
According to these figures the water from the glacier would have
filled the basin of Devils Lake to overflowing in about five years.
If the rate of advance of the ice was greater than assumed above
tT. C. Chamberlin and R. D. Salisbury, Geology, I, 261.
THE HISTORY OF DEVILS LAKE, WISCONSIN 365
or if the ice was thicker, and both postulates are reasonable, the
time required to fill the basin would have been less.
The foregoing figures may be roughly checked by an estimate of
the amount of material deposited by the glacial waters. The 6
miles of ice edge which drained into Steinke Lake furnished water
enough to deposit over 23 billion cubic feet of débris; in the Peck
basin water running from } mile of ice front deposited at least 142
million cubic feet of débris; it therefore seems safe to assume that
over 11 miles of ice edge furnished over 73 billion cubic feet of water.
The Devils Lake gap between the two moraine dams contains over
2 billion cubic feet of débris (10,000 feet in lengthX 1,000 feet in
width X 283 feet in depth) and most of this must have come in the
water from the two short stretches of glacier to the north and
southeast, the water from the Steinke and Peck basins doubtless
having been essentially clear.
It is not necessary to go farther to warrant the assumption that
Devils Lake must have risen during the glacial epoch until it
reached an outlet. On the edges of the basin there are only four
low points: (1) over the terminal moraine east of the south end of
the lake, (2) over or around the west edge of the moraine north of
the lake, (3) at the head of the south fork of Messenger’s Creek,
(4) at the head of the north fork of Messenger’s Creek. If the
openings to the east and north be considered to have been blocked
by the ice, as they doubtless were during the glacial occupancy,
the lowest outlet available was between the head of the north fork
of Messenger’s Creek and the head of a valley tributary to the east
fork of Skillett’s Creek, where the altitude is between 1,180 and
1,200 feet A.T.
Salisbury and Atwood’ found evidence that the glacial lake
stood go feet higher than the present lake, by finding erratic, iceberg-
floated bowlders in the talus on the west bluff of the lake at altitudes
of 1,050feet. Theorizing that the lake must have stood even higher
than this, that it must have had an outlet, and that icebergs would
float toward, and strand in, such an outlet, the writer has made
™R. D. Salisbury and W. W. Atwood, Bull. No. 5, Wis. Geol. and Nat. Hist. Surv.,
P, 133. ;
266 ARTHUR C. TROWBRIDGE
careful search for erratic bowlders in the valleys of the north and
south forks of Messenger’s Creek. An hour’s search revealed 103
such bowlders in the valley of the north fork, and an equal time in
the valley of the south fork failed to discover one. The highest
igneous rock bowlder in the north fork is at 1,162 feet, 202 feet
above present lake-level, and only 28 feet lower than the divide
across which the lake-water must have drained. Glacial cobbles
occur within 16 vertical feet of the divide, and one diabase cobble
was found on the west slope of the divide in the drainage of Skil-
lett’s Creek.
It is concluded, therefore, that during the Wisconsin epoch
the waters of Devils Lake stood against the glacier at the north
end, formed a bay up the valley to the northeast about as far as the
north-south road in the Peck flat, reached to about the level of
Elephant’s Rock on the east bluff, stood against the ice at the
southeast extremity of the basin, extended to within a short dis-
tance of the head of the south fork of Messenger’s Creek, and spilled
over the divide at the head of the north fork of Messenger’s Creek,
as shown in Fig. 1 and Fig. 4. Through this outlet the water
flowed west into a larger lake, known as the Upper Baraboo Lake,
now extinct.
THE POST-GLACIAL LAKE
As the edge of the glacier receded during the closing stages
of the Wisconsin glacial epoch, the high level of the lake and its
westward-flowing outlet may have been maintained for a time, but
when a connection was established between Devils Lake and a lake
which came into existence in the Baraboo valley, and whose surface
stood at a lower level, the waters of Devils Lake were lowered to the
lowest point in the morainic dams. The lowest point on the surface
of either moraine was a little east of the middle of the gap at the
north end of the lake, along the site of the railroad and wagon road
from the lake to Baraboo. The original level of this outlet is not
known, because it has now been cut almost to lake-level, but the
edges of the outlet gap on the surface of the moraine at either side
are at about 1,020 feet, or 60 feet above present lake-level. This
is approximately the level of the surface of Devils Lake after the
retreat of the ice,and before the outlet had been lowered appreciably.
THE HISTORY OF DEVILS LAKE, WISCONSIN 367
From the 1,020 foot level (60 feet above the present lake) the
lake surface probably was brought down rapidly by the lowering of
its outlet. The stream flowing north across the moraine cut its
SAN AWN
Rivet \
TAY Ko
BSS Sree Ait
7 NS
—<K
a
el x
AA
SA
N
ee NS WY MANS ve .
S =<S \ Y \ \ aR
Ns WS PG ekNORSOUANY
7 SONS
\ \ LN
WN
SY
= weNl
AN \
Bxy\ RWW
SEA r\ aes
‘a
~ |
SOS <S
\
NN
B MY \
INN WN
Ore! \)
RAYS
Sy
Fic. 4.—A map showing the shape and size of Devils Lake during the occupancy
of the ice. The shaded area shows the lake surface. The arrow near the left margin
designates the outlet. Scale: 1% inches equal 1 mile.
way downward in the glacial till, removing the fine material, but
leaving the bowlders on its bed, lowering the lake surface from 1,020
to about 970 feet.
368 ARTHUR C. TROWBRIDGE
The lake surface apparently stood at this level (970 feet) for a
time, perhaps because the outlet had made a resistant bed for itself
by the accumulation of bowlders on its bottom. The evidence of
this stage is a well-defined beach or barrier ridge of sand across the
north end of the lake, whose crest is 8-10 feet above the lake and
which confines a low, peaty area between it and the moraine. This
low area back of the ridge was clearly once a lagoon. The ridge
has been so strengthened artificially that it is impossible to tell
whether or not it was originally broken at the old outlet, but water
which could reach the top of the ridge could today flow north
through the old outlet to the Baraboo River.
During this second stage of the lake it is believed still to have
been receiving water from the Peck and Steinke basins (though the
lakes of these names were extinct) and from a later post-Glacial
lake northeast of the Steinke flat, which has become known locally
as Shubring Lake. Shubring Lake occupied what is now a flat
area, 1 mile by 4 mile in extent, 4 miles northeast of Devils Lake,
and just across the terminal moraine from the Steinke flat in the
area of ground moraine. The Shubring flat is bordered on the
north, west, and south by the inner edge of the terminal moraine
and on the east by a drift-covered hill of quartzite. The slopes
toward the flat are almost covered in a narrow belt parallel with the
edges of the flat by thousands of bowlders which almost form a
wall around the old lake bottom and which were concentrated on the
shores of the shallow lake by “ice push.” This lake was formed as
the edge of the ice retreated, leaving an inclosed basin. During the
first stages of recession of the ice, the lake received glacial waters,
and after the ice had left the confines of the basin precipitation
formed the source of supply. The line of outlet of the lake is plainly
seen as a flat-bottomed, bowlder-strewn, linear depression interrupt-
ing the course of the terminal moraine from the southwest end of
the Shubring flat to the Steinke flat south of it, and now used by
Mr. Shubring as a roadway across the terminal moraine. The
bottom of the valley which was the site of the outlet and the
bowlder wall around the lake flat are at the same altitude (hand-
level measurement), and not more than 2 feet above the level of
the flat. The original depth of the lake is unknown, no records
of the depth of the lacustrine fill being available.
THE HISTORY OF DEVILS LAKE, WISCONSIN 369
It is clear from a study of the post-Glacial drainage conditions
east of Devils Lake that the water from Shubring Lake and the
Steinke Basin continued to drain into Devils Lake by way of Peck
flat for a time at least after the withdrawal of the edge of the ice.
These conditions must have persisted until a tributary of the Bara-
boo River had time to work headward up through the ground
moraine on the slope of the South Range and through the terminal
Fic. 5—Sketch map showing the conditions of drainage around Devils Lake
during the first and second stages in the post-Glacial history of the lake. The endless
line marks the boundaries of the drainage basin of the lake during these stages.
moraine, to tap the Shubring and Steinke basins and divert their
drainage to the north, inaugurating present conditions. The
establishment of present drainage would require at least a time
commensurate with the time involved in the lowering of the outlet
of Devils Lake to the 970-foot level.
So long as Devils Lake had an outlet to the north, that is, until
the 970-foot stage was reached, the boundaries of its basin must
have been somewhat as shown in Fig. 5.
370 ARTHUR C. TROWBRIDGE
There remains but one step in the pre-human history of Devils
Lake. Its outlet to the north has been abandoned. The reasons
for the sinking of the surface of the lake below the level of the outlet
may be several. With the gradual establishment of drainage in the
ground moraine of the Baraboo valley, a tributary to the Baraboo
River worked its way headward up the drift-covered slope of the
South Range and through the terminal moraine into the northern
Fic. 6.—Sketch map showing the conditions of drainage around Devils Lake
after the diversion of water from the Steinke and Shubring basins and the abandon-
ment of the northward outlet. The endless line marks the boundaries of the drainage
basin of Devils Lake.
portion of the Steinke flat, diverting the drainage from the east
and northeast, which up to this time had gone to Devils Lake,
to Baraboo River. Working rapidly in the non-resistant material
of the high Steinke surface, the stream developed a tributary
which worked back through the terminal moraine, tapping the
Shubring basin west of its original outlet. This diversion of
drainage resulted in a considerable decrease in the supply of water
for Devils Lake (compare Figs. 5 and 6) and doubtless helped to
THE HISTORY OF DEVILS LAKE, WISCONSIN 371
cause the lake surface to sink below the level of its outlet. Con-
ceivably also the advance of the post-Glacial epoch was attended
by increasing temperature and increasing evaporation and by
decreasing precipitation, so that more water was lost by evaporation
than was supplied by precipitation. And perhaps the time came
when underground lines of drainage were established in the gravel
and sand of the drift, through which enough water was carried from
the lake to cause its surface to subside. Doubtless all these factors
and possibly others contributed to the lowering of the surface of the
lake and the abandonment of its outlet.
With the abandonment of the outlet, the stream from the Peck
basin, which had flowed into the lake or into its outlet, chose the
easier of two possible routes, avoiding the lake and flowing down the
valley of the old outlet to Baraboo. River. A few years ago, in
order to prevent floods in its lower course, this stream was diverted
again to Devils Lake by the building of a dam and the digging of a
shallow ditch connecting the stream with the lake. Today the stream
flows into Baraboo River or into Devils Lake, according as the
temporary dam is located in the stream channel or in the artificial
ditch.
The lake of today: has a maximum depth of only about 40 feet,
covers an area of only a little more than + square mile, and is without
an outlet. Its drainage basin at present is shown in Fig. 6.
SUMMARY
Devils Lake is seen to have had a long and complicated history.
(1) Pre-Cambrian rock formations were deposited and folded.
(2) Across the edges of the beds a peneplain probably was formed,
, over which a large stream meandered. (3) An uplift seems to have
occurred and the stream intrenched itself, making a deep, curved
gorge through a ridge of quartzite. (4) An early Paleozoic sea
advanced over a surface of high relief and great irregularity, par-
tially, but not completely, filling the pre-Cambrian gorge with
sediments. (5) This sea withdrew at the end of the Prairie du
Chien epoch, leaving a sag where the old gorge had been. (6) Condi-
tions favoring deposition in the sea, and perhaps deposition by wind
temporarily and locally, were renewed and deposition continued
272 ARTHUR C. TROW BRIDGE
until the gap was entirely filled and the pre-Paleozoic topography
was buried deeply. (7) The seas finally withdrew, stream courses
were developed and superimposed upon the old topography and
structure, and a peneplain was developed, probably in late Tertiary
time. (8) This peneplain was uplifted about 200 feet, and a second
partial peneplain was developed. During this erosion cycle the
Wisconsin River adjusted itself in the old gorge, and a valley about
300 feet deep was formed. (9g) Another uplift to an amount of
about 600 feet resulted in renewed erosion and the deepening of the
gorge from 300 feet to goo feet or more. (10) In the northern por-
tion of the renewed gorge so formed,an ice barrier lake was formed by
the edge of the Wisconsin glacier blocking the. valley at two points.
This lake received much glacial water, covered an area twice as large
as does the present lake, was at least 270 feet deep, and had an
outlet at 1,190 feet A.T., which drained northwestward. (11) With
the recession of the ice, the surface of the lake dropped to about
1,020 feet and then to 970 feet, as an outlet to the north was estab-
lished, and lowered through the terminal moraine dam. (12) Owing
to a diversion of drainage in the Steinke and Shubring basins,
decreasing the intake, and perhaps owing to changes in climate and
the establishment of underground channels, the surface of Devils
Lake fell from 970 to 960 feet, and the outlet was abandoned.
, | VN Prate IV
Vou. XX 0. 4
Bees OF GEOLOGY,
Jounal
SETTING OF Devits LAKE
: © TOPOGRAPHIC
Map SHOWING THE
. ical Survey
boo, Briggsville, Dells, and Denzer sheets of the United States Geologica
Taken from the Baraboo, Brigg 2 p
i
'f
Granite!
Aare
TARE TN
¥
Lie al
Huts
eT
wits
! :
NS
ieee
* Z t
Ore ay
m
a WORE)
een
i Nex
Mele
JournaL or GroLocy, Vou. XXV, No. 4
a
a
i
iit
: a
$ = 2. es ~e
oot Sao
Serre
=f oa Des
Iai — oe |
t Seeetine
2 sos Sen
a gorau |
°o (ero
TOCENE
ELaIS
VICTAN
EOLOGIC Map oF THE Vicinity OF Devyits LAKE
Volosit
Gr
OAMBRTAN
Rhyolice
PRE.
Panite
THE MIDDLE PALEOZOIC STRATIGRAPHY OF THE
CENTRAL ROCKY MOUNTAIN REGION
C. W. TOMLINSON
University of Chicago
PART III
STRATIGRAPHY—continued
THE SILURIAN SYSTEM
In Utah.—This system, not including the Richmond series, is
not known with certainty in the central Rocky Mountain region,
except in northern Utah, where it was included by Weeks" under
the name “Paradise limestone,’ which has been supplanted in
official usage by the “‘Laketown dolomite” of Richardson.” Its
fauna does not comprise many species, but is ample to demonstrate
that the system is Silurian. It is closely allied to the Niagaran
faunas of other regions.
In Nevada.—This system may be represented in the upper part
of the Lone Mountain limestone of western Nevada, although no
diagnostic Silurian fossils have been reported from that formation.
The following is an extract from Iddings’ section near Modoc Peak,
in the Eureka district:
3. Shaly limestone, rich in fossils. Lower part of Nevada limestone.
2. 550 feet. Light-gray siliceous limestone, with fine lines of bedding; in
upper portion weathering in almost rectangular fragments; growing less
siliceous toward the bottom.
t. 140 feet. Light-gray, highly crystalline, saccharoidal dolomite; not
siliceous.4
tF, B. Weeks, unpublished manuscript, U.S. Geol. Survey.
2G. B. Richardson, ‘‘The Paleozoic Section in Northern Utah,” Amer. Jour. Sci.,
4th Ser., XXXVI (1913), 406-15.
3 Cf. E. M. Kindle, ‘Occurrence of Silurian Faunas in Western America,’’ Am.
Jour. Sci., 4th Ser., XXV (1908), 125 fi.
4 Arnold Hague, of. cit., p. 66. Section measured by J. P. Iddings.
SiS)
374 C. W. TOMLINSON
These descriptions show a marked resemblance to the section
at Blacksmith Fork, from the lower part of the Jefferson down into
the Laketown dolomite.
On the south slope of Quartz Peak in the Pahranagat Range in
southern Nevada, about 140 miles south of Eureka, the Lone
Mountain includes the following member:
2. 335 feet. Massive bedded dark siliceous limestone, with a stratum (not
far above the base) 30 feet thick, almost made up of a species of Pentamerus.t
This, again, is strikingly like the Laketown.
Is the Laketown dolomite in part Devonian ?—The uppermost
member of the Laketown dolomite in the Blacksmith Fork section,
202 feet thick, is of much the same type as the Leigh formation
of northwestern Wyoming. It immediately underlies beds of
typical Jefferson dolomite. In the Teton River section the basal
member (23 feet thick) of Blackwelder’s? Darby (Jefferson) forma-
tion is of similar character, and is separated by an erosion surface
from the underlying Leigh formation. In the Livingston Peak
section there is a member, 21 feet in thickness. which is identical
in type with the true Leigh, but which lies above the cliff-making
Upper Bighorn dolomite, at the base of the Jefferson dolomite.
In the Crandall Creek section the corresponding member, 26 feet
thick, overlies a 47-foot sequence of variegated beds which lie
disconformably upon the Upper Bighorn. At Livingston Peak and
at Teton River, ostracods like those which are characteristic of the
Leigh formation were found in this repetition of the Leigh type '
at the base of the Devonian system.
In brief, the uppermost member of the Laketown dolomite in
northern Utah corresponds in lithologic character, and in relation
to the overlying Jefferson dolomite, to the member which farther
north appears to have been the introductory deposit of the first
‘Devonian submergence. This relation suggests that the Utah
member in question belongs with the Devonian rather than with
the Silurian system. This interpretation has been followed in the
correlation tables and diagrams accompanying this thesis, where the
beds just discussed appear as Member 2 of the Devonian system.
t Arnold Hague, op. cit., p. 196.
2 Eliot Blackwelder, unpublished manuscript, U.S. Geol. Survey.
PALEOZOIC STRATIGRAPHY OF ROCKY MOUNTAINS 375
Their much greater thickness in Utah may mean that the Devonian
‘submergence proceeded slowly northeastward from that region, or
that deposition was more rapid there.
fe)
Q
{
[ug
O
|
O
O
g900' (7)
Pre-Laketown and pre-Devonian emergence.—The base of the
Laketown dolomite at Blacksmith Fork is clearly disconformable
upon the Fish Haven formation. No evidence of a break between
Silurian and Devonian has been noted in that section; and Kindle,
Fic. o—Map showing the extent and thickness of the Silurian system
376 C. W. TOMLINSON
chiefly because of the persistance of Halysites catenulatus into the
Jefferson dolomite, favors the idea of continuous sedimentation in
this locality from Silurian into. Devonian. Elsewhere in the Rocky
Mountains, however, where the Silurian is absent, there was cer-
tainly an emergent interval immediately preceding the inauguration
of Devonian sedimentation. The fact that no physical evidence
of hiatus at that time has been noted as yet in northern Utah means
little. The recurrence of Halysites in the Devonian certainly
means no more, as regards continuity of submergence, than its
persistence from the Richmond into the Silurian; yet the discon-
formity between the Richmond and the Silurian on Blacksmith
Fork is well marked.
How far the Silurian system originally extended over the Rocky
Mountain province can only be conjectured. Throughout Wyo-
ming, wherever the Devonian occurs, it is apparent that the Silurian
system, if ever represented, was completely removed Oy erosion
prior to the Devonian submergence.
A small fauna collected by Blackwelder in the Gros Ventre
Range from beds just below the Leigh dolomite was referred by
Kindle and Weller to the Silurian, and by Ulrich to the Richmond.’
The latter interpretation is probably correct, as the Leigh of the
Teton Range is confidently correlated with part of the Richmond
series in the Bighorn Range. Blackwelder’s section in the Gros
Ventres does not include beds of the types which characterize the
Laketown dolomite in Utah. |
The Silurian is not represented in Hintze’s? section in the central
Wasatch. The beginning of known Devonian deposition in that
region was certainly preceded by an interval of emergence. Neither
Silurian nor Devonian strata are known in the Uinta Range.’
THE DEVONIAN SYSTEM: THE JEFFERSON DOLOMITE
The basal division of the Jefferson dolomite-—The age of the
beds here called Members 1 and 2 of the Devonian system has
been discussed under the Silurian. Member 3, which follows 2 in
1 Eliot Blackwelder, “Origin of the Bighorn Dolomite of Wyoming,” Bull. Geol.
Soc. Amer., XXIV (1913), 610.
2 Op. cit. 3 F. B. Weeks, op. cit., XVIII (1907), 427-48.
PALEOZOIC STRATIGRAPHY OF ROCKY MOUNTAINS 377
apparent conformity wherever both are present, was the earliest
development of the most characteristic Jefferson type of dolomite
—dark-brown in color, saccharoidal, and giving forth a strong
*SUOT]EUIO] pa}yeai0d pue (UBTUOADG d9I[PPII
pue [zg] Jomoy) o}IWO[Op Uosiayef[ 9} JO ssouyoIy? pue jUus}xo oy} Burmoys deyy—‘or ‘ory
OdvyuO TO)
f
pe : {
: Soe Le i VQVAIN
SF
\ sv SS iar !
4 DUON, i 7 7 i 44)
! j—_ > nl
j { { 001 | TO a ee =
! \ \ ‘
| SN |WO AVM NG, j
uon? y p 2 ~. i
) N Quon? ) 9° to, ie j
(,LS1 420 2°N) j2uon Ze ue EN O oo | V Gg | j
(,L1€ AVA YON) (i+) LET ) set? | ae
4 j . (
\
\
|
i IEE 4aAo JON
|
{
{
The data suggesting dis-
conformity at the top of Member 3 are summarized in the dis-
bituminous odor when freshly broken.
cussion of disconformities (pp. 131-34).
378 C. W. TOMLINSON
The main division of the Jefferson dolomite is dominated by a
rather uniform succession of fairly massive, fetid, dark-brown
dolomites, but includes toward its base a minor variegated sequence
of thin members (4 to 8, inclusive), among which are light-colored
~ dolomite, and locally (Labarge Mountain, Blacksmith Fork) sand-
stone. The correlation of these minor members on a lithologic
basis has been attempted, but cannot be considered very reliable.
In no two localities is there the same succession of members
throughout the formation, although there is a notable degree of
correspondence between the Crandall Creek and Teton River sec-
tions, for example. In the Dead Indian Creek, Livingston Peak,
Logan, Labarge Mountain, and Blacksmith Fork sections the
variegated members are relatively ill-developed, and the out-
standing feature of the formation is a nearly uniform sequence,
200 feet thick or more, of what has been described above as the
most characteristic Jefferson type of dolomites.
Members 12 to 14 of the Devonian system, above the main
body of the Jefferson dolomite, are differentiated only at Blacksmith
Fork and at Labarge Mountain, where they have a combined thick-
ness exceeding 200 feet. They appear not to be represented in the
sections in northwestern Wyoming and southern Montana. This
fact, with certain other evidence, is strongly suggestive of an
emergent interval between Members 14 and 15. (See discussion of
disconformities, p. 132.)
The upper division of the Jefferson dolomite—Member 18 is
distributed with remarkable uniformity over a very wide area,
and has been distinguished in nearly every locality within the scope
of this thesis where the Jefferson dolomite has been described in
any detail. Its brecciated structure, together with an abundance
of calcite geodes in some places, give its weathered surfaces a char-
acteristic nodular, pitted appearance. It is everywhere separated
from the main mass of the Jefferson by a sequence of thin-bedded,
largely platy dolomites. At Labarge Mountain the base of this
sequence is marked by ro feet of sandstone (Member 15); and in
the Teton River section its lower portion contains some thin beds
of sandstone, and quartz grains are scattered through several of the
beds of dolomite.
PALEOZOIC SERATIGRAPHY OF ROCKY MOUNTAINS 379
This upper division of the Jefferson, as indicated by the correla-
tion diagrams, maintains a fairly constant development even where
the main body of the formation beneath it shows much variability.
The beds doubtfully referred to Member 18 in the Goose Creek
Ridge and Rattlesnake Mountain sections are not typical of that
member, but resemble it in being massive and in underlying a
series of thin-bedded dolomites which possess all the essential
characters of Member 19. (See discussion of the Three Forks
formation, pp. 383-84.) It is perhaps more likely that the beds in
question really belong to the Richmond series.
Misuse of the name “ Jefferson” in Yellowstone Park and vicinity.
—The name “ Jefferson’’ was first used by Peale’ to describe the
dark limestones overlying the Gallatin formation in the Three
Forks quadrangle, Montana. ‘The strata included by Peale under
that name are now believed to be entirely of Devonian age As
they comprise everything between the Gallatin formation below
and the Three Forks shale above in the type locality of the forma-
tion, it was natural for other geologists to apply the name “ Jeffer-
son”’ to all the strata between those two formations in neighboring
areas. In the Livingston quadrangle, Montana, and in the Ab-
saroka* quadrangles, Wyoming, however, Hague confined the
name “‘ Jefferson”’ to the strata which he regarded as Silurian, and
extended the name “Three Forks” to include all the limestones
carrying Devonian fossils. Iddings and Weed, in Yellowstone
Park,’ employed the names in similar fashion, but not altogether
consistently in different parts of the Park.
With the aid of his own field notes, and by as careful a correla-
tion of members as the published descriptions permit, the writer
1A. C, Peale, “The Paleozoic Section in the Vicinity of Three Forks, Montana,”
U.S. Geol. Survey, Bull. 10 (1895).
2 EK. M. Kindle, letter of March 29, 1916; Edwin Kirk, letter of June 13, rors.
3 Arnold Hague, ‘Description of the Livingston Sheet,” Geol. Atlas U.S., Folio 1
(1894).
4 Arnold Hague, “Description of the Absaroka Quadrangle,” Geol. Ailas U.S.,
Folio 52 (1899).
5J. P. Iddings and W. H. Weed, “‘ Descriptive Geology of the Gallatin Moun-
tains,” U.S. Geol. Survey, Monographs, XXXII, Part 2 (1899), chap. i; ‘‘ Descriptive
Geology of the Northern End of the Teton Range,” ibid., chap. iv; W. H. Weed,
“Geology of the Southern End of the Snowy Range,” ibid., chap. vi.
380 C. W. TOMLINSON
has prepared the accompanying tentative correlation table (Fig. 11)
of the sections measured by Iddings and Weed, adding three of his
own sections for comparison. The “ Jefferson” of Hague, Iddings,
and Weed is seen to include the Bighorn dolomite as its chief
RIVER | CREEK
(tdding s X Want iee Iddings &|(Iddings & ffiddings &
Division
(cr) No.
150” }measured,
Mo (22)
: 175°) 180°
7 42-43A (at)
Sei bl va 180°
2 i (29)
200°| 62°
(19)
@l re (o
MADISON | FoRMATION
TRRE
FORKS
100°
nN
pe
i]
On
baer
9
DEVONIAN
JEFFERSON
4
o>
a
»
ee
~
—_
ad
Three
BIGHRORN
Ae
WwW =
sd
ORDOVICIAN
Zz
q
2
«
w
2
te}
I
4
uw
ne =
a
>
ra
uw
3
roy
J
Bae
Conventions same as in Fig. 2. Page numbers at Top of columns
vefer To pages in U.S Geot. Survey Monograph XXX, Part 2.
Fic. 11.—Tentative correlation table for sections in Yellowstone Park and vicinity
constituent, and their “Three Forks limestone”’ consists chiefly of
beds which are properly to be correlated with part of the true
Jefferson dolomite as originally defined by Peale. The Devonian
fossils identified by Girtyt from Yellowstone Park came from beds
1G. R. Girty, “Devonian and Carboniferous Fossils of the Yellowstone N ational
Park,” U.S. Geol. Survey, Monographs, XXXII, Part 2 (1899), chap. xil.
PALEOZOIC STRATIGRAPHY OF ROCKY MOUNTAINS 381
well down in the Jefferson formation, which explains their likeness
to the Jefferson fauna known from other regions.t_ The variability
of the Three Forks formation from shale to limestone along the
strike has been overestimated because of this same erroneous
correlation.
Correlation of the Jefferson dolomite with the Nevada limestone.—
The Jefferson dolomite, as known in Montana and northern Utah,
has been correlated definitely by Kindle? with the Nevada limestone
of eastern Nevada, on paleontological grounds. Of the eleven
specifically identified forms described by Kindle from the Jefferson
formation, other than new species, seven were described by Walcott
from the Nevada limestone. Of the 32 forms partially or com-
pletely identified by Kindle from the Jefferson Limestone, only
three are of genera not described by Walcott from Nevada.
The total thickness of the Nevada limestone is estimated by
Hague’ at 6,000 feet. Although accuracy is not claimed for this
figure, it cannot be doubted that the Nevada is very much thicker
than any described section of the Jefferson limestone. The follow-
ing figures are given by Hague for the thickness of the Nevada
limestone at various localities in the Eureka district:
Newark Mountains (Bed 5): 3,500 feet. Considered by Hague® to be
less than (the upper) half of the total thickness of the formation. A small
Upper Devonian fauna was collected “several hundred feet below the top.”
Near Modoc Peak? (Beds 3-18): 4,710 feet. (Hague regarded the Nevada
as including more than 700 feet of lower strata also.) Rich Lower Devonian
fauna in lower 425 feet (Beds 3-4); no fossils other than “‘Stromatopora” and
“Chaetetes”’ found at higher horizons.
East of Lamoureux Canyon’ (Beds 1-5): 3,000 feet, top not exposed.
Basal 200 feet carries a rich Lower Devonian fauna.
County Peak:9 4,500 feet. Fossils at three horizons.”
«Cf. E. M. Kindle, “Fauna and Stratigraphy of the Jefferson Limestone in the
Northern Rocky Mountain Region,” Bull. Amer. Pal., 1V, No. 20 (1908), 22.
2 Ibid., pp. 20-21.
3C. D. Walcott, “Paleontology of the Eureka District,” U.S. Geol. Survey,
Monographs, VIII (1884).
4 Arnold Hague, “Geology of the Eureka District, Nevada,” U.S. Geol. Survey,
Monographs, XX (1892), 13, 63-64.
s Ibid., pp. 82, 158. 7 Ibid., p. 66. 9 Ibid., p. 68.
6 [bid., p. 158. 8 [bid., p. 67. 7 Tbid., pp. 78-80.
382 C. W. TOMLINSON
The occurrence of such great thicknesses at four distinct local-
ities renders it unlikely that the true thickness has been greatly
overestimated because of duplication by faulting.
The two chief fossiliferous horizons in the Nevada limestone are
near the base and near the top, and are separated by from 2,000 to
4,000 or more feet of strata which have yielded no diagnostic fossils.
The fauna described by Kindle from the Jefferson limestone includes
10 forms’ (2 of which are specifically identified) which occur only in
the lower part of the Nevada limestone, 4? (3 of which are spe-
cifically identified) which occur in both the upper and the lower
horizons, and 53 (2 of which are specifically identified) which occur
in the upper part of the Nevada only. Although this evidence is
meager, it suggests that the Jefferson limestone includes represent-
atives of both the basal and upper parts of the Nevada limestone.
It is possible that the disconformity between Members 14 and 15
of the Jefferson, for which evidence is cited on page —, represents
a large part of the barren middle portion of the Nevada. This
would mean that the greater thickness of the Nevada, as com-
pared with the Jefferson, was due, at least in part, to the presence
in the Nevada of members which either never were deposited in the
Rocky Mountain province or were eroded from that region in
Devonian time; that during the Devonian period the eastern part
of the Great Basin was more persistently submerged or less ele-
vated in emergent intervals (or both) than was the central Rocky
Mountain Region.
In the Pahranagat Range, west of Hiko, in southeastern Nevada,
Walcott measured a section including 5,400 feet of strata which he
assigned to the Devonian. From this sequence he obtained 22
fossil forms, to 9 of which he assigned definite specific names. All
of these forms are identical, according to Walcott’s lists, with
t Favosites sp., Chonetes cf. macrostriata, Stropheodonta sp., Schuchertella che-
mungensis, Athyris sp., Platyceras sp., Actinopteria sp., Loxonema approximatum ?,
Loxonema nobile, Bythocypris ? (Leperditia?) sp.
2 Stromatopora sp., Productella cf. spinulicosta (Productus subaculeatus), Atrypa
reticularis, Martinia maia.
3 Spirifer utahensis Meek (S. disjunctus Sowerby), S. engelmanni Meek, Ptert-
nopecten sp., Naticopsis sp., Pleurotomaria sp.
4 Arnold Hague, op. cit. ult., pp. 197-99.
PALEOZOIC STRATIGRAPHY OF ROCKY MOUNTAINS 383
species found in the Nevada limestone in the vicinity of Eureka—
some in the lower, some in the upper, fossiliferous horizon there.
THE DEVONIAN SYSTEM: THE THREE FORKS FORMATION
Type secttons.—As defined by Peale’ in the type sections in the
Three Forks quadrangle, Montana, this formation includes all
strata between the top of Member 18 of the Jefferson dolomite and
the base of the Madison limestone (Mississippian). Raymond?
has collected and studied large faunas from several horizons in the
type locality opposite Logan. Haynes? has carried on Raymond’s
work, and studied in detail the stratigraphy of the formation in all
exposures within about twenty miles of Logan.
Members 19 and 20.—The chief fossiliferous zone of the Three
Forks, Member 20 of the Devonian system (Members 4 and 5 of
Haynes) has not been identified in Montana east of the Three
Forks quadrangle, nor anywhere in Wyoming. It is quite possible
that outliers of it may yet be found in those areas, however, as
Member 1g is widely distributed there. The top of the Jefferson
dolomite is clearly marked in nearly every section in western
Wyoming and southwestern Montana by the massive, brecciated
Member 18. Were it not for this, the beds of Member 19 might
easily be confused with those of Member 17, to which they bear
much likeness. In the Dead Indian Creek section, where Member
18 was not recognized, certain strata were assigned to Member 19
because their relation to overlying strata corresponds to the rela-
tion of Member ro to overlying beds in the Crandall Creek section,
where Member 18 is present.
Beds of doubtful age in the Bighorn Range, and near Cody.—In
the Goose Creek Ridge and Rattlesnake Mountain sections there
is a belt of platy buff and yellow dolomites and calcareous shales
t A.C. Peale, ‘‘The Paleozoic Section in the Vicinity of Three Forks, Montana,”
U.S. Geol. Survey, Bull. r10 (1893).
2P, E. Raymond, ‘‘On the Occurrence in the Rocky Mountains of an Upper
Devonian Fauna with Clymenia,”’ Amer. Jour. Sci., XXIII (1907), 116-22; “The
Fauna of the Upper Devonian of Montana,” Annals of the Carnegie Museum, V (1909),
141-58; ‘“‘The Clymenia Fauna in the American Devonian,” Proc. Seventh Intern.
Zool. Cong. (Boston, 1907).
3 W. P. Haynes, op. cit., pp. 13-54.
384 C. W. TOMLINSON
between the top of the fossiliferous Bighorn and the base of the
typical Madison, which corresponds closely in lithological character
with Member 1g as developed farther northwest. In the Bighorn
Se
COLORADO
?)
)
ocecs enn
ee a) =
eee Se
o_o Fe cares cease soa mame 2 ae 0 ear
“109°(?)
@ (None
]
i,
None j
91032’ (Benson limesta
?
ay
‘= . eserr 7 a
Sececesboor*
e
0
Sncce
c-)
s
ie
e
oe
a et
Range, where this belt is 238 feet thick, it was considered by Darton’
to be the basal member of the Madison limestone; but it is unlike
«N. H. Darton, ‘Description of the Bald Mountain and Dayton Quadrangles,”
Geol. Atlas U.S., Folio 141 (1906), p. 4; Goose Creek section.
Fic. 12.—Map showing the extent and thickness of the Three Forks (Upper Devonian) and
correlated formations.
PALEOZOIC STRATIGRAPHY OF ROCKY MOUNTAINS 385
any part of the true Madison seen by the writer elsewhere. So far
as known, it is wholly barren, as is likewise Member 19. It may
belong to the Richmond series. However, if the Devonian system
is represented in the sections named, it includes this belt, and pos-
sibly the underlying massive member which separates it from the
fossiliferous Upper Bighorn.
Member 21, comprising black shales overlain by yellow or red
aranaceous strata, is found at Crandall Creek and Dead Indian
Creek in the Absaroka Range with a development almost precisely
like that shown in most of Haynes’s sections near Three Forks.
Haynes' considers that the fauna of this member is transitional to
the Mississippian.
The Three Forks formation in northern Utah; correlation with
the Benson and Ouray limestones—Richardson* reports very poor
exposures of the Three Forks formation in the Randolph quadrangle
in northern Utah, with red shaly limestone (Member 19 ?) as the
most conspicuous element in the float. The fauna collected by
him from this formation, and identified by Kindle, indicates a
mingling of Ouray and Three Forks species. Five? of the 7 species in
Richardson’s list occur also in the Devonian fauna of the Ouray
limestone in Colorado,4 and 65 of the 7 occur in the Three Forks
formation of the type area.
Haynes® has noted the probability of a correlation between the
Three Forks and the lower (i.e., the Devonian) part of the Ouray.
Of 25 species and varieties of brachiopods listed by Kindle? from
the latter horizon, and of 27 listed by Haynes® from “Bed 5” (part
of Member 20 of this paper) of the former, there are 11 which occur
in both lists.
1 Op. cit., pp. 21, 28. 2 Op. Cit., D. 412.
3 Camarotoechia cf. contracta, Productella coloradensis, Schizophoria striatula var.
australis, Spirifer whitneyi var. animasensis, and Spirifer notabilis.
4E. M. Kindle, “‘The Devonian Fauna of the Ouray Limestone,” U.S. Geol.
Survey, Bull. 391 (1909).
5 C. contracta, P. coloradensis, S. striatula var. australis, S. whitneyi var. anima-
sensis, Syringothyris cf. carteri, Cleiothyridina sp.
6Op. cit., p. 24. |
7E. M. Kindle, ‘‘The Devonian Fauna of the Ouray Limestones,”’ U.S. Geol.
Survey, Bull. 291 (1909), p. 12.
8 Op. cit., p. 25.
386 C. W. TOMLINSON
Richardson' estimates the thickness of the Three Forks forma-
tion in the Randolph area at 200 feet. In the Blacksmith Fork
section both the upper and the lower limits of the Three Forks are
marked by the usual sharp lithologic contrasts, yet the strata
between the Jefferson and the Madison total 978 feet in thickness.
For the most part, this sequence is composed of the usual Three
Forks types; but between the green shales below and the black
shales above there appears a 360-foot belt of blue-gray limestones
(Member 20B).
Less than ten miles south of the above section, in the canyon
east of Paradise P.O., Utah, Kindle? noted no representative of the
Three Forks formation. In the region east of Ogden such a repre-
sentative is probably to be found, as suggested by Richardson,3
in the reddish shales and thin-bedded limestones mentioned by
Blackwelder,* which are 250 feet thick on the South Fork of Ogden
River.
In the South Fork of Big Cottonwood Canyon, about 75 miles
south of Blacksmith Fork, occurs the Benson limestone of Hintze,5
comprising 1,032 feet of blue limestone. From a horizon in the
upper part of this formation Hintze collected a fauna which is very
like that of the Ouray limestone. His. description of the Benson
indicates that it is lithologically very different from the typical
Three Forks. The belt of blue-gray limestones in the middle of
the Three Forks formation at Blacksmith Fork, and the upper
division of the Jefferson dolomite in the same section, correspond
roughly in character with the Benson. This fact, together with the
mingled Ouray—Three Forks fauna found in the Randolph quad-
rangle, suggests the possibility that the Ouray lithologic type
dovetails into the Upper Devonian rocks of the more northern and
western province.
«G. B. Richardson, op. cit., p. 412. ;
2E. M. Kindle, “‘The Fauna and Stratigraphy of the Jefferson Limestone in the
Northern Rocky Mountain Region,” Bull. Amer. Pal., IV, No. 20 (1908), 16.
3 Op. cit., p. 412. Also Eliot Blackwelder, letter of April 29. 1915.
4 Eliot Blackwelder, ‘‘New light on the Geology of the Wasatch Mountains,
Utah,” Bull. Geol. Soc. Amer., XXI (1910), 528-20.
5 F, F, Hintze, Jr., op. cit., pp. 88-142.
PALEOZOIC STRATIGRAPHY OF ROCKY MOUNTAINS 387
SUMMARY OF RESULTS ON SPECIAL PROBLEMS OF
INVESTIGATION
The following progress has been made toward the solution of
the four special problems noted in the foreword of this thesis:
1. The Age of the “‘ Jefferson dolomite” in the Livingston, VY ellow-
stone National Park, and Absaroka quadrangles.—It has been shown
beyond doubt that the Ordovician Bighorn dolomite is present, and
characteristically developed, throughout the Absaroka Range, as
far north as Livingston Peak in Montana, and in Yellowstone
Park. It was included in the “Jefferson dolomite” of Hague,
Iddings, and Weed. It is followed disconformably by the true
Jefferson dolomite (Devonian), much of which was erroneously
included by earlier writers under the name “Three Forks forma-
tion.”
2. Middle Paleozoic disconformities—A well-marked discon-
formity between the Lower (Trenton) and Upper (Richmond)
divisions of the Bighorn dolomite has been discovered in the Ab-
saroka Range. In northwestern Wyoming and _ southwestern
Montana the existence of disconformities at the base of the Big-
horn dolomite and at the base of the Jefferson dolomite has been
established. Much evidence has been gathered which points
toward the existence of disconformities at at least one horizon within
the Jefferson dolomite, and at the base of the Mississippian system.
3. Correlation of Upper Cambrian and Ordovician formations in
Wyoming with those in Utah——The Upper and Lower divisions of
the Fish Haven dolomite in northern Utah have been shown to cor-
respond in lithologic character to the Upper and Lower divisions,
respectively, of the Bighorn dolomite; and there is evidence of a
disconformity between the two divisions of the Fish Haven dolo-
mite, as well as between those of the Bighorn. Several members
of the Gallatin formation of Wyoming and Montana have been
tentatively correlated on lithologic grounds with members of the
St. Charles formation (Upper Cambrian) in northern Utah, and
the main flat-pebble conglomerate zone at the top of the Gallatin
is tentatively correlated with a member of similar character in
Utah which includes the top of the St. Charles formation and the
lower part of the Garden City (Beekmantown) formation.
388 C. W. TOMLINSON
4. Relation of Ordovician to Silurian in Utah.—In the vicinity
of Blacksmith Fork, northern Utah, the Laketown dolomite
(Silurian) has been found to be separated from the underlying
Upper Fish Haven dolomite (Richmond) by a well-marked dis-
conformity.
HISTORICAL SKETCH
UPPER CAMBRIAN AND LOWER ORDOVICIAN
Extent of submergence.—The present extent and variations in
thickness of the sediments deposited in the central Rocky Mountain
region during the Upper Cambrian-Lower Ordovician submergence
are shown in Fig. 7._ It will be noted that this is not a paleogeo-
graphic map in the usual sense, as no attempt has been made to
indicate the original extent of these sediments, other than to show
them as continuous across relatively small areas, from which they
have certainly been removed by post-Cretaceous erosion.
The much greater thickness of this group of sediments in western
and northern Utah and in Nevada, as compared with Wyoming
and Montana, means either that (1) sedimentation continued
longer or (2) took place more rapidly in the first-named region, or
that (3) during the succeeding interval of emergence erosion removed
a larger part of the series in question from Wyoming and Montana
than from the adjoining region to the southwest. Certainly the
first and last, and perhaps all three, of these factors combined to
produce the net result. A logical supposition is that the sea
retreated from northeast to southwest, so that deposition continued
in the Great Basin after erosion had begun on recently emerged
areas in Wyoming and Montana. The wide-spread uniformity
in type of the sediments in question indicates a similarly extensive
uniformity of conditions of deposition, and suggests that the second
factor was not of great importance.
Paucity of clastic sediments—The beginning of the Upper
Cambrian transgression in Utah was marked by a deposit of sand
of varying thickness. From that time until the close of the Beek-
mantown epoch no part of the region in which the St. Charles,
Garden City, Gallatin, and Upper Deadwood formations are found
PALEOZOIC STRATIGRAPHY OF ROCKY MOUNTAINS 389
received any notable quantity of arenaceous sediment, so far as the
record now shows. During long intervals the deposits accumulat-
ing on the floor of the clear sea which covered this area were nearly
or quite without clastic ingredients. At no time was clastic mud
brought in in such quantity as greatly to preponderate over cal-
careous matter. This condition makes it certain that there were
no high shores, nor debouchures of streams draining areas of high
relief or dry climate, in or very near to the region under consid-
eration.
Evidence that the sea was shallow.—The intraformational con-
glomerates which characterize the upper part of the Upper Cam-
brian and much of the Beekmantown series record the fact that the
seas of that time were shallow enough to permit the breaking and
movement, by waves or currents, of freshly or partly consolidated
sediments on the sea bottom.
The Chazyan(?) sands.—Somewhat later in the Ordovician
period, probably in the Chazyan epoch, a flood of sand swept over
the extreme northern part of Utah and much of eastern Nevada
(see Fig. 8). This was probably the result of an uplift of neigh-
boring lands, perhaps in central or eastern Utah.
Emergence and peneplanation.—Subsequently, the entire region
shown in the accompanying maps (Figs. 6-10, 12, 13) emerged and
the freshly exposed sediments were subjected to extensive erosion,
which continued until about the close of the Black River epoch.
This emergence took place without any appreciable deformation of
the Cambrian and early Ordovician sediments. Probably it pro-
duced no great relief in the region where those sediments are known.
The first deposits of the succeeding submergence were laid down on
a surface whose relief probably did not exceed 200 feet in all north-
ern and western Wyoming, and on which, in that region, slopes
as steep as 10° were, so far as known, nowhere more than a few feet
or yards in length. Either the relief was not greater than this at
any time during the interval of emergence or a rougher surface was
reduced to this condition by stream erosion during that interval,
with the aid of marine planation at the border of the advancing
Middle Ordovician sea.
390 ' CC. W. TOMLINSON
MIDDLE AND UPPER ORDOVICIAN
The Trenton submergence.—About the end of Black River time
the sea readvanced into the central Rocky Mountain region,
probably from the west. The area known to have been covered
by this inundation in Utah and Wyoming is approximately the
same as in the case of the Late Cambrian submergence (see Fig. 9).
Locally, in Wyoming a little sand was deposited in depressions of |
the surface as the sea-border transgressed the region; but the
main body of limestone which represents this Trenton submergence
is wholly free from arenaceous and shaly matter. The seas were
even more persistently clear than in Upper Cambrian time.
The masses of calcareous alge which make up much of the
Lower Bighorn and Lower Fish Haven dolomites bear witness
to the prevalent shallowness of the water.
The post-Trenton emergence.—Sedimentation was interrupted by
emergence between Trenton and Richmond times, but the erosion
accomplished during this interval seems to have been slight. If
any considerable thickness of strata was removed at this time, no
remnants of them have been recognized. It is probable that the
land was not uplifted much above the base level of the streams which
drained its surface.
The Richmond submergence.—With the return of the sea in the
Richmond epoch the processes of deposition were resumed under
conditions similar to those which obtained during the Trenton
epoch. The seas remained clear until the end of the Ordovician
record. As in the case of the Upper Cambrian formations, the
Bighorn and Fish Haven dolomites present no evidence, aside from
the sandstone locally found at the base of the Bighorn, of proximity
to shore lines. The original extent of the Bighorn formation was
probably much greater than the area in which it now isfound. The
long period of emergence which followed its deposition afforded
ample opportunity for erosion.
SILURIAN
Emergence.—During the early part of the Silurian period
(Medina and Clinton epochs) the entire central Rocky Mountain
region was above sea. The emergence, like the two or three next
PALEOZOIC STRATIGRAPHY OF ROCKY MOUNTAINS 391
preceding ones, was accomplished, so far as the known record
shows, without crustal deformation. In Wyoming, Montana, and
Colorado, this interval of emergence may have continued without
interruption until Devonian time.
The Niagaran( ?) invasion.—Toward the middle of the period,
probably in the Niagaran epoch, the sea invaded this region again,
from the Great Basin side. The sediments deposited in this sea
are known only in northern Utah (and in Nevada ?), but may once
have been far more extensive (see Fig. 10). The Conchidium
knighti fauna found in the Laketown dolomite implies a marine
connection with Alaska and with Europe.
DEVONIAN
Pre-Jefferson erosion.—The Jefferson dolomite overlaps far
beyond the Laketown dolomite into western Wyoming and south-
western Montana. In Wyoming it rests upon various horizons
of the Upper Bighorn dolomite. In Montana, north and west of
Livingston, it lies directly upon the Upper Cambrian. It is prob-
able that the Bighorn dolomite once extended farther into Montana,
and was removed by pre-Jefferson erosion. If the Silurian system
originally extended into Wyoming or Montana, it also must have
been removed at this time. In Utah, no evidence of a hiatus
between the Silurian and the Devonian has been noted.
Although the Jefferson dolomite overlaps both the Silurian and
the Ordovician systems, the surface upon which the first sediments
of the Jefferson dolomite were deposited appears to have been
almost as level as that on which the Lower Bighorn dolomite was
laid down. A suggestion of relief is furnished by the contrast
between the sections at Logan and Livingston, Montana. Near
Logan, the Jefferson rests upon Member 3 of the Cambrian system.
At Livingston Peak, about 50 miles to the east, 438 feet of Bighorn
dolomite intervene between the Jefferson and that member. Even
this difference indicates but an inappreciable slope.
Devonian marine invasions.—The Devonian submergence, dur-
ing which the Jefferson dolomite was deposited, was probably
interrupted in Wyoming and Montana by at least one interval of
emergence. ‘The sea appears to have made two or three successive
392 C. W. TOMLINSON
advances from Nevada northeastward, so that the thickness of its
deposits diminishes toward the northeast because of both a pro-
gressively shorter total duration of sedimentation in that direction
and, correspondingly, a progressively longer duration of the inter-
vening emergences, with accompanying erosion. A small amount of
sand, diminishing toward the northeast, was deposited in the shore-
ward portion of the sea during each advance. Aside from this, little
or no clastic material was laid down in the Devonian sea in the
central Rocky Mountain region until the later part of the period.
The Upper Devonian muds.—In the Upper Devonian the cal-
careous sediments accumulating on the sea bottom in Utah, Wyo-
ming, and Montana (see Fig. 12) were polluted by an influx of mud,
which in Utah attained a thickness of several hundred feet. The
Three Forks formation comprises the resulting shales and lime-
stones, together with a small thickness of arenaceous sediments,
only locally present, which may date from the beginning of the
Mississippian submergence rather than from the close of the
Devonian. The sea which occupied parts of Colorado and central
Utah in Upper Devonian time seems to have been connected
directly with the Three Forks sea for a short time only, near the
middle of the epoch.
THE PRE-MISSISSIPPIAN INTERVAL OF EMERGENCE
Depth of erosion.—It is certain that all of Wyoming, most of
Montana, and much of Colorado was above sea-level and under-
going erosion at the close of the Devonian period. The Lower
Mississippian sediments (Madison limestone) overlap all older
Paleozoic formations (see Fig. 13). The gradual truncation of
the Bighorn dolomite and the upper part of the Deadwood forma-
tion by the Madison limestone in the southern part of the Bighorn
Range is clearly due to pre-Mississippian erosion, and the varying
thickness of the Three Forks formation in northwestern Wyoming
and part of southwestern Montana is due to the same cause. The
relief of the final surface on which the Madison limestone rests,
however, is nowhere very pronounced.
Where the Mississippian strata rest upon Ordovician rocks a
part of the hiatus may be attributed to erosion during Silurian and
PALEOZOIC STRATIGRAPHY OF ROCKY MOUNTAINS 393
early Devonian time; and where they lie on the Cambrian, the pre-
Bighorn emergence also must be reckoned with. Again, it is
probable that parts of the region never were covered by strata of the
uorsai ureyunoyy AYIOY [ero oy} ur snjzery werddississrpy-ord ay} Jo yIZue] oy} Surmoys depw— £1 “oly
upiluoaaq] uO \reqeung- 24g \ : |
\uo Burasal } { upjuoaag saddy ‘ !
¢) ajqeuuojuo7 \ stozosayq) , ()
eee
12
ee.
ueiddississijqj yadda pue
C)
\uerddississy) |
urjuode] sadd) fijso7z
o-
ssogeecr
U
ueiddississ Hay
t)
WW ,
UseMmjaq snyeipy
of
£
@
t
a
By
q
3
PG
i sueiuoAaq taddy 10)
Ss {
oo puUOWY IY] UBaMmjaq sNyeiL] \
Oe 0 ‘
me AS?
ee UeiddissisCy,
- o “SIA pue 'y
4 (¢) udiuode] 2/PP!W \
uaamjaq snjzei
JW 19q vwH i
athe
is certainly not
The other extreme,
extreme view, that none of the existing Middle Paleozoic formations
ever extended much beyond their present limits,
age of some of the formations which appear elsewhere.
true in the central Rocky Mountain region.
304 C. W. TOMLINSON
that every one of them once extended over the whole area and was
cut back by erosion before the Mississippian period, is likewise,
in all probability, incorrect.
“ Positive’’ and ‘‘ Negative” areas; axes of warping.—Whatever
was the true proportion between erosion and non-deposition, the
net result is shown in Fig. 13, which is, in effect, a paleogeologic map
of the central Rocky Mountain region at the close of the pre-
Mississippian interval of emergence. If the area of Cambrian
and pre-Cambrian rocks which then extended from eastern Utah
northeastward across southern Wyoming and northern Colorado
owed its exposure to the erosion of Middle Paleozoic formations
during emergent intervals, it certainly was more exposed to erosion
during those intervals than were the adjacent areas on either side
of it—probably because of greater altitude. If its exposure was
due wholly to non-deposition of the Middle Paleozoic formations,
then it must have been above water when neighboring areas were
submerged. In either case, or under any hypothesis which com-
bines the two views, this area of Cambrian and pre-Cambrian rocks
must have been at a greater average altitude during Middle Paleozoic
time than the areas on either side of it.
The major axis of this area of upward tendency is approximately
at right angles to the average trend of the Rocky Mountain folds. The
downward-tending belt, which is known to have been covered by
most of the Middle Paleozoic seas, extends in parallel fashion from
the northern part of the Great Basin northeastward across the
Rocky Mountain province. This is well illustrated by the distri-
bution of the Bighorn dolomite (see Fig. 9).
The trend of these belts indicates the direction of axes of warping
merely, not of true orogeny.
The pre-Mississippian interval of emergence was brought to a
close by an invasion of the sea, whose deposits are more widespread
in the central Rocky Mountain region than are those of any preced-
ing submergence.
REVIEWS
Mineral Deposits of the Santa Rita and Patagonia Mountains,
Arizona. By FRANK C. SCHRADER, with contributions by
James) M>Hirn. U.S:. Geol. Survey, Bull. 582, tors. Pp.
373) Pls. 25.
This report covers an area of some 1,400 square miles extending
north and east from the city of Nogales, Arizona, on the Mexican border.
The mountain groups are short, irregular ranges bordered by broad,
sloping plains of Quaternary gravels. Mesozoic granite, quartz diorite,
and quartz monzonite, granite porphyry and <aplite, with Tertiary
andesite, rhyolite, and bedded tuffs and agglomerates, constitute the
two groups of igneous rocks which outcrop over about two-thirds of the
mountainous area. The remaining third is occupied by a thick sequence
of Cretaceous (?) shales unconformably underlain by Carboniferous
and Devonian limestones and older shales and quartzites.
The chief mineralization in this area accompanied or followed the
Mesozoic intrusions. Silver, copper, gold, and lead are the leading
metals. Fissure veins, contact-metamorphic deposits, replacement
deposits, and gold placers are all of notable importance. Copper is the
most important metal in the contact deposits, while silver and lead
dominate in the replacement type, which is here rather carefully dis-
tinguished from the contact deposits proper. The average depth of
ground-water level and of the oxide zone is about 250 feet.
CoW.E.
Some Mining Districts in Northeastern California and Northwestern
Nevada. By JAmMes M. Hii. U.S. Geol. Survey, Bull. 594,
TOLS.. Ep. 1096, pls. 12:
This is a report on a reconnaissance of nineteen widely scattered
mining districts below first rank in present-day importance, sixteen of
them in Nevada and three in California The chief value of the work
will be to those who are specially interested in the development of one or
more of these districts, though a number of points of general interest
in economic geology are brought out.
395
396 REVIEWS
Gold-silver, silver-lead, copper, and antimony deposits are described.
Two epochs of mineralization are recognized, the first following the
intrusion of a group of granitoid rocks in late Cretaceous or early Ter-
tiary time, the second following the extrusion of a group of later Ter-
tiary andesites and rhyolites. Most of the ores are believed to date
from the former epoch. The gold-silver and silver-lead ores are vein
deposits. Most of the copper ores are replacements of sedimentary
rocks in the vicinity of monzonitic intrusives.
A number of good examples of magmatic differentiation are described.
CuWe t
‘““A Geologic Reconnaissance of the Cuzco Valley, Peru.” By
HERBERT E. GrReGorY. Am. Jour. Sci., XLI (1916), pp.
1-100, pls. 2, figs. 44. Contributions from the Peruvian
Expedition of 1912 under the Auspices of Yale University and
the National Geographic Society.
The bottom of Cuzco Valley ranges from 10,000 to 11,000 feet
above the sea. The higher peaks of its bordering ranges reach heights
of from 14,000 to nearly 16,000 feet. Between 13,000 and 13,500 feet
above the sea, there is a well-defined plateau surface, called by Gregory
the “Inca peneplain.”” It is characterized by much gentler slopes than
those of the deep canyons which have been incised into it by the present
streams. It is underlain for the most part by complexly folded and
faulted sedimentary rocks. Quaternary glaciation molded all of this
region above 13,500 feet, and occasional moraines descend to 12,500
Teets
The Cuzco Basin, occupying the upper end of the valley, is about
ten miles long by three miles wide. It is believed to owe its origin to
faulting. It is floored with lacustrine sediments, probably of pre-glacial
age, which are overlapped by large alluvial fans, now being dissected.
At the head of the valley is a plateau surface sloping from 11,500 feet
up to 13,000 feet above sea-level, underlain by Cretaceous limestones
carrying marine fossils—the only marine sediments known in this
region. The great bulk of the mountain masses surrounding the valley
is made up of clastic sedimentary rocks: brown sandstones, chocolate
shales, and conglomerates. These are divided by Gregory into three
formations, one of which is composed chiefly of pyroclastic material.
They are tentatively assigned to the Permian and Jura-Trias. There
REVIEWS 307
are also mapped several bodies of intrusive diorite and syenite, and small
areas of andesitic and basaltic lavas, all doubtfully assigned to the
Tertiary.
Cu Wet;
Boone County. By C. E. Kress and D. D. TEETs, Jr. West
Virginia Geol. Survey, 1915. Pp. 648, pls. 52, figs. 3, maps 2.
County reports now published cover the greater part of the northern
and western sections of the state. In these reports are chapters on
physiography and mineral resources, but those treating of the stratigraphy
of the area are of more general interest.
In Boone County the outcropping rocks range in age from the middle
of the Conemaugh to near the base of the Kanawha series. The Kana-
wha has a remarkable development. It has been, differentiated into
29 formations totaling 1,844 feet in thickness. About 30 coal beds
from 1 to 15 feet thick are intercalated in the series. Scores of partial
sections are given.
There is a preliminary report on the paleontology of the county,
and an excellent geologic map accompanies the report in a separate
COVEY.
_W. B. W.
Guidebook of the Western United States. Part B, The Overland
Route. By W.T. LEE, R. W. Stone, H.S. GALE, and OTHERS.
U.S. Geol. Survey, Bull. 612, 1915. Pp. 244, pls. 50, figs. 20,
maps 25.
This series of guidebooks is without question the best ever published
and should find a wide use among the traveling public of the United
States. This volume serves at once to direct the attention of the traveler
to the things most worth observing in the land through which he passes,
and to render more interesting every stage of the journey. Even the
best-informed person, who has been over the route many times, cannot
fail to profit by the use of it, and those planning a trip for the first time
will find in it by far the most complete, reliable, and attractively written
guide available. A wealth of historical, geographical, and geological
information is woven together into an interesting and comprehensive
whole, written in narrative style. The industries and agricultural and
mineral resources of the regions passed through are discussed, and a few
398 REVIEWS
appropriate statistics are given. Withal, the book is not too long to be
easily read in the course of the journey.
Although technical terms are consistently aveided! with the excep-
tion of a few essential ones which are explained in a brief glossary,
a large amount of geological information of general interest is included.
The numerous photographs are exceptionally well chosen, and well
adapted to awaken interest in geology. There are 25 admirable maps
on a scale of 1:500,000, showing topographic, geologic, and cultural
features, and mounted in a manner convenient for the reader.
This bulletin covers the route followed by the Union Pacific from
Omaha to Ogden, that of the Southern Pacific from Ogden to San Fran-
cisco, and that of the Oregon Short Line from Ogden to Yellowstone
National Park. It is obtainable from the Superintendent of Documents,
Washington, D.C., for fifty cents, postage free.
CW. 2:
Gold on the North Saskatchewan River. By J. B. TYRRELL. Cana-
dian Mining Inst., Toronto, 1915, pp. 68-81.
Summarizes the general geology of the region, and describes the
occurrence of gold in the gravels of the stream. The gold is said to be
most abundant from Goose Encampment to Beaver Lake Creek, a dis-
tance of 130 miles. Some gold has been recovered from gravel taken
out for use on the streets of Edmonton.
A. -Di4B
Die mikroscopische Untersuchung der Erzlagerstatten. By Georg
Berc. Berlin, 1915. Pp. 108, figs. 88. 7
A book for use in the laboratory. The work is divided into four
parts, as follows: (I) optical and microchemical methods, covering
opaque and transparent minerals, reactions for the identification of
compounds and elements, chemically and by means of anlauf farben;
an appendix deals with manipulation, separation, and preparation of
material, etc.; (II) microscopic characters of the more important ore
and gangue minerals; in this section the minerals discussed are grouped
according to crystal system; in addition to their appearance under the
microscope, the more important physical characters are given. Asso-
ciated minerals are usually mentioned; (III) the microscopic structure
on the important types of ore deposits; a large number of figures illus-
trate typical sections of the various kinds of deposits; the grouping is
REVIEWS 399
genetic and mineralogic; (IV) petrography of the thermally and pneu-
matolytically altered rocks.
The work should be of great value, as it is the most complete that
has come to the reviewer’s hands, and the field is one of great and growing
importance. As a pioneer work it will doubtless need additions in the
near future, but for the present it fills a long-felt want.
AnD B:
Mineral Resources of New Mexico. By Fayette A. Jones. Bull.
1, State School of Mines Mineral Resources Survey of New
Mexico. Socorro, 1915. Pp. 77, map ft.
A catalogue of the various mineral products that the state is thought
to be capable of producing. The entire lack of statistics of production
detracts from the value of the book, and the potential possibilities of
the state in the matter of resources not now being exploited is left largely
to the optimism of the reader.
A. Di B.
A Gold-Platinum-Palladium Lode in Southern Nevada. By ADOLPH
Knorr. U.S. Geol. Survey, Bull. 260-A. Washington, 1915.
Pp. 1s, pl: er.
Describes briefly the occurrence of a gold-platinum-palladium vein
in dolomite in the Boss mine in the southern point of Nevada. The
claim was first worked for copper, as the gold, occurring in black particles,
was not readily recognizable as such. The veins are strongly oxidized
and present development does not reveal conclusive evidence as to the
origin or the original mineralogical character of the vein. The deposit
apparently adds a new type to the list of American ore deposits. Values
up to Pt. 99 oz., Au. 111 oz., and Pd. 16 oz. are reported.
A. D. B.
Guidebook of the Western United States. Part C, The Santa Fe
Route. By N. H. Darton, and OTHERS. U.S. Geol. Survey,
Bull. 613, 1915. Pp. 194, pls. 42, figs. 40, maps 24.
This bulletin is in all respects a mate to the preceding, organized
in the same effective way. The many simple structure sections which
accompany this work, and which give even the most casual reader a fair
conception of the geologic structure along the route, are especially to
be commended.
400 REVIEWS
It is to be hoped that the demand for these guidebooks will be suffi-
cient to warrant their revision and republication from time to time, so
that they may be kept thoroughly up to date. It is suggested that in
future editions a set of small maps of the individual states be added, to
give the reader a more complete geographic background for the detailed
route maps than is afforded by the single physical map of the United
States which is included in each volume.
Bulletin 613 covers the route of the Atchison, Topeka & Santa Fe
Railway from Kansas City, Missouri, to Los Angeles, California, with
a side trip to the Grand Canyon of the Colorado. It‘is obtainable from
the Superintendent of Documents, Washington, D.C., for fifty cents,
postage free.
Cr We
“Diamond Fields of German South-West Africa.” By C. W.
BotsE. The Mining Magazine (London), June, 1915, pp.
1-14, figs. 8.
These fields, which produced nearly $15,000,000 worth of diamonds
in the year 1913, constitute a narrow strip along the coast in the south-
western part of the colony. The productive area is about 75 miles
long, nowhere extending more than 12 miles inland. It is a barren
desert, swept by the south trade winds, and has an annual rainfall not
exceeding 2 inches. The coastal tract is characterized by low north-
south ridges separated by stretches of sand and fine gravel, in which the
diamonds are found. The productive stratum is at the surface, and
averages not more than six inches in depth. The deposits are in effect
eolian placers, in which the diamonds have been concentrated by the
sifting action of the wind, which winnows out the finer and lighter mate-
rial. The richest concentrations are often found in streaks parallel to
the direction of the prevailing wind. The average size of the diamonds
found increases toward the southern part of the field, but their original
source is unknown.
CA Wrelt:
Coal Fields of Pierce County. By JosEpH DANIELS. Washington
Geol. Survey, Bull. 10, 1914. Pp. 146, pls. 30, figs. 23.
The coals of this county are all in the Puget formation of Eocene age.
This formation consists of sandstones, shales, and coals and attains a
remarkable thickness, estimated at 15,000 feet. The beds have been
REVIEWS 401
sharply folded and faulted. A heavy mantle of glacial drift covers most
of the county, and details of structure are learned only from mine work-
ings. No estimate is made of the amount of coal available.
W. B. W.
Geology and Water Resources of Tularosa Basin, New Mexico. By
O. E. MEINZER and R. F. Hare. U-S. Geol. Survey, Water
Supply Paper 343, 1915, pp. 317, pls. 19, figs. 51.
The Tularosa Basin is shown to be a down-faulted block between
highlands of Cretaceous, older Mesozoic (?), and Carboniferous sedi-
mentary rocks lying upon granite. The Pennsylvanian Manzano group
of Red Beds here has a thickness of about 2,500 feet, and contains much
gypsum. Tertiary intrusives of several types cut the older rocks. The
valley bottom is covered with Quaternary deposits, comprising water-
laid gravels and finer sediments several hundred feet thick, together
with modern dune sands and saline deposits. There are two recent lava
flows, with well-preserved cinder cones and craters.
An unusual feature of this valley is an area of 270 square miles of
dunes of gypsum sand, still in motion. The gypsum is derived from
deposits on the floor of a large alkali flat to windward (west) of the dune
area. The gypsum of the playa in turn was derived from the bedded
gypsum in the Manzano group, the solution of which has given rise to
numerous sink-holes, locally so abundant as to have produced karst
topography. aoa
Limestone Road Materials of Wisconsin. By W. O. HOTCHKISS
and EDWARD STEIDTMAN. Wisconsin Geol. Survey, Bull. 34,
Lords) Rp. 126, pls. 41, ngs. 2. | :
The importance of thorough investigation of road-building materials
is shown by the fact that this state appropriated approximately $1,250,-
ooo for highway purposes in 1914. This report treats of limestone
materials only. Part I describes various standard tests on road mate-
rials and emphasizes the importance of thorough testing. The chief
limestone horizons are discussed briefly. Part II takes up by counties
the limestone areas of the state. There is a brief description of lime-
stone resources, with results of samples tested, and 4o areal geology
maps of different counties. Wisconsin is said to be more abundantly
supplied with road materials than any of the neighboring states.
W. B. W.
402 REVIEWS
Logan and Mingo Counties. By R. V. HENNEN and D. B. REGER.
West Virginia Geol. Survey, 1914. Pp. 776, pls. 15, figs. 23,
maps 2.
The 1914 contribution to the excellent series of county reports of
this state includes two counties on the southwestern border. The
general treatment is similar to that of earlier reports.
The strata exposed range from middle Pottsville to the lower mem-
bers of the Conemaugh series. A large number of detailed sections of
these series are given. A table of 150 coal analyses, both proximate and
ultimate, is given, and under separate cover is a map showing areal and
economic geology and structure geology.
W. B. W.
Biennial Report of Vermont State Geologist. By G. H. PERKINS
and OTHERS. 1913-14. Pp. 448, pls. 78, figs. 41.
The greater part of this report treats of the marble industry of dhe
state. It contains reprints of Bulletins 521 and 589 of the United States
Geological Survey which deal with commercial marbles of this area.
Separate articles by various writers give brief résumés of the geology
and mineralogy in the vicinity of Hardwick, Woodbury, and Benning-
ton. The talc deposits of the state are described by E. C. Jacobs. He
believes that these deposits have resulted from the SEH gEI SI of
basic intrusions into sedimentary country schists.
W. B. W.
Biennial Report of Missouri Bureau of Geology and Mines. By
H. A. BUEHLER. 1913-14. Pp. 62. 5
This is chiefly an administrative report of work completed by the
survey staff during this biennial period. Statistics on mineral produc-
tion in the state during 1913 and 1914 also are given.
W. B. W.
Devonian of Southwestern Ontario. By C. R. STAUFFER. Geol.
Survey of Canada, Geol. Series, No. 63. Pp. 341, pls. 20,
map I.
Devonian beds outcrop over the entire area of the Ontario province
that projects southwest between Lakes Huron and Erie. Probably the
entire system is present although the correlations of the upper and
REVIEWS 403
lower members are still tentative. The beds rest unconformably on
Silurian rocks ranging in age from Salina to Cobleskill or younger.
The revised classification follows:
{Port Lambton (probably Portage or Chemung)
\Huron shale (probably Genesee)
(ite Ipperwash limestone
: Petrolia shale
Hamilton) Widder beds
Olentangy shale
Delaware limestone
eal limestone {Onondaga limestone
\Springvale sandstone (local facies)
Oriskany sandstone
Lower Helder (wanting or possibly represented in the
Detroit River series)
Upper
Devonian
From 2 to ro sections are given in each of the 12 counties included
in the area. The paleontology has been worked out with great care and
each section is accompanied by its fauna classified by horizons. Of
the species given, 350 are listed from the Hamilton beds alone and 347
from the Onondaga. Largely on faunal evidence the Springvale sand-
stone is considered a local facies of the Onondaga instead of belonging
to the Oriskany.
There is a chapter that summarizes present knowledge of the devel-
opment and migrations of the Devonian faunas in this region and a
chapter of bibliography on the Devonian of the eastern continental area.
W. B. W.
Central Connecticut in the Geologic Past. By JOSEPH BARRELL.
Connecticut Geol. and Nat. Hist. Survey, Bull. 23. Pp. 44,
figs. 9.
This bulletin is a study of the extent to which ancient geologic
structure and physiographic features may be reconstructed from data
now available. Technicality has been avoided in an attempt to make
the report available for general reading. To further this plan the his-
torical geology is taken up in reverse of the usual order.
A number of wholly new structure sections are of chief interest.
These sections reproduce the structure for each geologic period since
late Paleozoic times. There is a departure from conventional structure
sections in that clouds and the landscape of the background are added.
These features may be of aid to readers untrained in geology.
404 REVIEWS
Many of the generalizations on the great events of geologic history
apply to a much larger area than that named in the title of the report.
W. B. W.
Peat Resources of Wisconsin. By F. W. HUELs. Wisconsin Geol.
Survey, Bull. 45, 1916. Pp. 274, figs. 20, pls. 22.
As neither oil, natural gas, nor coal is found in this state, special
interest attaches to its peat deposits, and has resulted in their system-
atic examination. Part I contains a general discussion of the origin
of peat, its preparation and uses. Part II gives a description of the
state’s peat deposits. These are limited to the drift-covered area. The
quantity of peat land is placed between two and three million acres and
the amount of peat between two and three billion tons. Analyses show
that for the most part the peat compares favorably in quality with peats
now being used extensively in Europe.
A number of companies have engaged in peat production but all
have suspended operations. It seems probable that the greater part
of the peat lands will be drained and reclaimed for agricultural purposes.
W. B. W.
Soils of Mississippi. By E. N. Lowe. Mississippi Geol. Survey.
Pp) 220) figsi) 22)
This is a preliminary dicussion of the subject and is to be followed
later by a complete report. The state is divided into g soil areas that
correspond roughly to physiographic provinces. The soil of chief
geologic interest is found in a belt of loess, that extends the length of the
state, and borders the Mississippi River flood plain.
An appendix to the report contains a number of soil analyses.
W. B. W.
Hudson Bay Basins and Upper Mississippi River. U.S. Geol.
Survey, Water-Supply Paper, 355, 1915.
This volume is one of a series of twelve reports of measurements of
stream flow in the United States during 1913. The data cover the flow
of the larger streams in Minnesota draining into Hudson Bay and those
of Minnesota and Wisconsin that are tributary to the Mississippi.
W. B. W.
THE PRINCIPLES OF STRATIGRAPHY
By AMADEUS W. GRABAU, S.M., S.D.
Professor of Palaeontology in Columbia University -
“This monumental work will be of great value especially to the
younger generation of American Geologists, in broadening their
mental horizon and outlining the problems which rise for solution
in the twentieth century study of the rocks. It differs from other
manuals in the English language to such a degree that it
supplements but does not supplant them.””—Science, July 14, 1914.
Buo; 1150 pages, 264 illustrations. Price $7.50
DESCRIPTIVE CIRCULAR SENT UPON REQUEST
A. G. SEILER & CO., NEW YORK CITY.
: Cambridge | Aniversity Dress
An Introduction to Geology. By J. E. Marr, Sc.D., President of the Geological
Society. With 33 figures. Crown 8vo. $1.00.
The Principles of Stratigraphical Geology. By J. E. Marr, Sc.D.,F.R:S., F.G.S.
With 25 figures. Crown 8vo. $1.60. Cambridge Geological Series.
The Geology of the Lake District, and the Scenery as influenced by Geological Struc-
ture. By J. E. Marr, Sc.D., F.R.S., Fellow of St. John’s College, Cambridge. With
a map, frontispiece, and 51 illustrations. Demy 8vo. $3.60.
Agricultural Geology. By R. H. Rastatt, M.A., late Fellow of Christ’s College and
Demonstrator of Geology in the University of Cambridge. With 51 illustrations.
Demy 8vo. $3.25. Cambridge Geological Series.
Petrology for Students: an introduction to the study of rocks under the microscope.
By ALFRED Harker, M.A., E.R.S., F.G.S., Fellow of St. John’s College. Fourth edition,
revised. With gi figures. . Crown 8vo. $2.00. Cambridge Geological Series.
A Treatise on Crystallography. By W. J. Lewis, M.A., Professor of Mineralogy in
the University of Cambridge, Fellow of Oriel College, Oxford. With 553 figures. Demy
8vo. $4.50. Cambridge Geological Series.
Published by the Cambridge University Press, England
G. P. Putnam’s Sons, Agents in the United States
2, 4and 6 West 45th Street, New York City
GC entecn Pagads
of Satisfaction
Direct to You by Parcel Post
REMINGTON JUNIOR TYPEWRITER
A labor saver for the home and an educator for
the children.
A work facilitator and thought accelerator for
professional men.
A time saver and business safeguard for the
farm and office.
REMINGTON TYPEWRITER COMPANY
(Incorporated)
374 Broadway, New York
Send me a Remington Junior Typewriter, price $50,
on free examination. It is understood that I may re-
turn the machine, if I choose, within ten days. If I
decide to purchase it, I ance to pay for it in ro
monthly payments of $5 eac
FINE INKS 4x ADHESIVES]
For those who KNOW
q radi ey
/ Drawing Inks
Eternal Writing Ink
By f Engrossing Ink
Le 2° 9 Taurine Mucilage
i | | g gin 8 Photo Mounter Paste
Drawing Board Paste
: Liquid Paste
Office Paste
Vegetable Glue, Etc.
Are the Finest and Best Inks and Adhesives
Emancipate yourself from the use of corrosive and
ill-smelling inks and adhesives and adore the
Higgins Inks and Adhesives. They will bea
revelation to you, they are so sweet, clean, well
put up, and withal so efficient.
At Dealers Generally -
CHAS. M. HIGGINS & CO., Mfrs.
Branches: Chicago, London
When a Projection Lantern
is to. be Purchased
—then careful consideration should be
given to the many advantages, the wide
range of usefulness and the moderate price of the
Bausch lomb
BALOPTICON
THE PERFECT STEREOPTICON
The beautifully sharp and clear images of the Balopticon are of the greatest advantage in educational work.
Scientific and durable in construction, superior in optical equipment—and with several models possessing the
great advantage of the new eascailed| Mazda lamp which gives increased illumination at less current expense.
Models for lantern slides from $25 up. Models for projecting opaque objects (maps, photos, specimens, etc.),
$35 up. Combined models with instant interchange from slide to opaque objects, $45 up. Model C (illustrated)
is of the slide type, with 400-watt Mazdalamp. Price $42
Our interesting illustrated price list with Fall descriptions sent free on request
Bausch ¢F lomb Optical G.
554 ST. PAUL STREET, ROCHESTER, N.Y.
NEW YORK WASHINGTON CHICAGO SAN FRANCISCO
Leading American Makers of Microscopes, Projection Apparatus, Photo, Sree Lenses, Ophthalmic Lenses and
nstruments, Stereo-Prism Binoculars, and other High-
tade Optical Products.
271 Ninth Street Brooklyn, N.Y. |
eee ee
zB
a's
2
Fe,
=
ses:
VOLUME “Saitee NUMBER 5
= ‘) 7 THE
JOURNAL or GEOLOGY
_A SEMI-QUARTERLY
: EDITED BY
b. THOMAS C. CHAMBERLIN AND ROLLIN D. SALISBURY
Je : With the Active Collaboration of
SAMUEL W. WILLISTON, Vertebrate Paleontology ALBERT JOHANNSEN, Petrology
STUART VABIEISS Invertebrate Paleontology ROLLIN T. CHAMBERLIN, Dynamic Geology
ALBERT D, BROKAW, Economic Geology
ASSOCIATE EDITORS
WS
SIR ARCHIBALD GEIKIE, Great Britain JOSEPH P.IDDINGS, Washington, D.C.
» CHARLES BARROIS, France JOHN C. BRANNER, Leland Stanford Junior University
_ ALBRECHT PENCK, Germany RICHARD A. F. PENROSE, Jr., Philadelphia, Pa.
~ HANS REUSCH, Norway F *WILLIAM B. CLARK, Johns Hopkins University
GERARD DEGEER, Sweden - WILLIAM H. HOBBS, University of Michigan
‘T. W. EDGEWORTH DAVID, Australia FRANK D. ADAMS, McGill University
BAILEY WILLIS, Leland Stanford Junior University CHARLES K. LEITH, University of Wisconsin
- GROVE K. GILBERT, Washington, D.C. WALLACE W. ATWOOD, Harvard University
' CHARLES D. WALCOTT, Smithsonian Institution WILLIAM H, EMMONS, University of Minnesota
ig HENRY S. WILLIAMS, Cornell University ARTHUR L. DAY, Carnegie Institution
*Deceased,
“— ) JULY-AUGUST 1917
_ THE LAWS OF ELASTICO-VISCOUS FLOW Se er ee ON ACN Cuen cone oe
yi ae PHILOGENY AND CLASSIFICATION OF REPTILES = - §. W.Wittiston grr
OUR PRESENT KNOWLEDGE OF ISOSTASY FROM GEODETIC EVIDENCE
wes WILLIAM BowlE 422
_ THE SATSOP FORMATION OF OREGON AND WASHINGTON - J Haren Bretz 446
“THE CORROSIVE ACTION OF CERTAIN BRINES IN MANITOBA = —_R.C. Wattace 450
i NOTES ON THE 1916 ERUPTION OF MAUNA LOA - - - - Harry O. Woop 467
ek. PROPOSED DIP PROTRACTOR - = ees - CHESTER K. WENTWORTH 480
~ PETROLOGICAL ABSTRACTS AND REVIEWS - - - = - ALBERT JOHANNSEN 402
N RECENT PUBLICATIONS - - - - 2 < 2 bs & fi a u os
aaa
es
Be ~ THE- UNIVERSITY. OF CHICAGO PRESS
CHICAGO, ILLINOIS, U.S. a
AGENTS:
THE CAMBRIDGE UNIVERSITY PRESS, Lonpon anp EDINBURGH o
THE MARUZEN-KABUSHIKI-KAISHA, Tokyo, Osaka, Kyoto, Fukuoka, SENDAI
THE MISSION BOOK COMPANY, SHANGHAI
Mais Yee
Neciyay
THE JOURNAL OF GEOLOG a
EDITED BY
THOMAS C. CHAMBERLIN AND ROLLIN D. SALISBURY
With the Active Collaboration of
SAMUEL W. WILLISTON ALBERT JOHANNSEN
Vertebrate Paleontology ie Petrology
STUART WELLER ROLLIN T. CHAMBERLIN
Invertebrate Paleontology Dynamic Geology
x
a
So
ALBERT D. BROKAW
_ Economic Geology
The Journal of Geology is published by the University of Chicago at the University Press semi-
quarterly, on or about the following dates: February 1, March 15, May I, June 15, August i September 15,
November 1, December 15. {The subscription price is $4.00 per year; the price of single copies is
65 cents. Orders for service of less than a half-year will be charged at the single-copy rate. {] Postage is
prepaid by the publishers on all orders from the United States, Mexico, Cuba, Porto Rico, Panama Canal
Zone, Republic of Panama, Hawaiian Islands, Philippine Islands, Guam, Samoan Islands, Shanghai.
{| Postage is charged extra as follows: For Canada, 30 cents on annual subscriptions (total $4.30), on
single copies, 4 cents (total 69 cents); for all other countries in the Posta] Union, 53 cents on annual
subscriptions (total $4.53), on single copies, 11 cents (total 76 cents). [Patrons are requeSted to make
all remittances payable to the University of Chicago Press in postal or express money orders or bank
drafts.
The following agents have been appointed and are authorized to quote the prices indicated:
For the British Empire: The Cambridge University Press, Fetter Lane, London, E.C.4. Yearly
subscriptions, including postage, 18s. 8¢. each; single copies, including postage, 3s. 2d. each.
For Japan and Korea: The Maruzen-Kabushiki-Kaisha, 11 to 16 Nihonbashi Tori Sanchome,
Tokyo, Japan. Yearly subscriptions, including postage, Yen 9.00 each; single copies, including
f
postage, Yen 1.50 each.
For China: The Mission Book Company, 18 Peking Road, Shanghai. , Yearly subscriptions, —
$4.00; single copies, 65 cents, or their equivalents in Chinese money. Postage extra, if mailed direct
outside of Shanghai, on yearly subscriptions, 53 cents, on single copies, 11 cents.
Claims for missing numbers should be made within the month following the regular month of
publication. The publishers expect to supply missing numbers free only when losses have been sustained
in transit, and when the reserve stock will permit. Deliveries are not guaranteed within the war zones of
Europe, nor in other countries where transportation is disturbed by conditions incident to war.
Business correspondence should be addressed to The University of Chicago Press, Chicago, Ill.
Communications for the editors and manuscripts should be addressed to the Editors of THE
JOURNAL OF GEOLOGY, the University of Chicago, Chicago, IIl.
Contributors are requested to write scientific and proper names with particular care and in citations
to follow the form shown in the pages of the Journal.
Separates, if desired, must be ordered in advance of publication. Separates of leading articles will be
printed as ordered, of which 50 (without covers) will be furnished graézs, the actual cost of the remainder
(and covers, if desired) to be paid for by the author. The actual cost may vary, and will depend upon
the amount of work in re-making the pages into forms, presswork, paper, binding, etc. Separates
containing half-tones may be expected to cost somewhat more, the increase depending upon the number ~
of cuts and the amount of work required upon them.
Entered as second-class matter, March 20, 1893, at the Post-office at Chicago, IIl., under the Act of March 3, 1879
)
efi
ies
eee
Pgs
ap
PELE SE DES ORAL BEGET RE ase = i .
~
hn Si tee HE
es
ot
> ae
VOLUME XXV NUMBER 5
THE
MOURNAL OF GEOLOGY
VULV AUGUST Tor,
THE LAWS OF ELASTICO-VISCOUS FLOW
A. A. MICHELSON
University of Chicago
When a solid is subjected to a strain beyond the “elastic limit,”
its behavior may be summarized as follows:
First: The application of the stress results in a rapid elastic
yield which, if inertia be negligible, is practically instantaneous.
If the stress be now removed, the specimen returns to its former
position.”
Secondly: This is followed by a slower yielding whose rate, if
the stress is not too great, diminishes with time and which ulti-
mately attains a constant value which may be zero.
If the stress be now removed, the specimen returns almost
instantaneously to a point short of its original position and then
1 The term ‘‘elastic limit” is very vague and should be replaced by limits which
may be characterized as follows:
a) The first limit is that within which the specimen returns instantly to its original
zero.
Beyond this first limit, if stress be instantly removed, the specimen promptly
returns to a position short of its original one, which we may designate as the “‘new
zero.”
b) The second limit is that beyond which the specimen does not return to its
original position or to the ‘‘new zero” even after a long time.
c) The third limit is that value of the stress which produces rapid yielding or
rupture. ;
2 In many cases the time interval between application and release of stress cannot
be made sufficiently short for complete, instantaneous recovery.
405
406 A. A. MICHELSON
continues at a much slower rate and ultimately comes to rest at
a point short of its original position.
If the stress is too great, the slow yield may increase until
rupture occurs."
The following may be considered as a provisional attempt to
formulate the behavior of substances under stress by the simplest
expressions which have been found to satisfy all the essential
requirements.
The formulae which follow are, in fact, sufficiently general to
cover every case thus far examined, including materials of widely
different properties, such as lead, tin, copper, aluminum, zinc,
iron, steel, quartz, glass, calcite, limestone, slate, marble, wax,
pitch, gelatin, and rubber. It may, however, be expected that
a more thorough investigation will require modification in the
formulae which may be made to fit special cases with greater
accuracy.
The type of strain selected for this investigation is the torsion
of cylindrical rods, as this is the only strain in which the form
remains unaltered. It is very probable that the laws governing
this special type may be made to include other distortions, such
as extension, compression, bending, etc.
Very decided changes may be expected from the effects of
temperature and pressure,” but these may be taken into account
by an appropriate alteration in the value of the “constants” which
enter into the formulae.
« Rupture may occur in consequence of such slow yielding, or it may be prac-
tically instantaneous. In the former case the result is due to separation of the viscous
coupling; in the latter, to the snapping of the spring.
2 A preliminary investigation of the effect of hydrostatic pressure on elasticity
and on viscosity was begun several years ago. It was hoped that this would show
results in conformity with those which maintain in the body of the earth—whose
enormous pressure produces an increase in both rigidity and viscosity sufficient to
make the body of the earth (which at its actual temperature under ordinary condi-
tions would certainly be in a molten state) as solid as steel. This expectation has
been partially realized for a number of materials, metallic and non-metallic, the results,
notwithstanding certain anomalies—traceable to the effects of previous history—
showing a perceptible increase in rigidity and a very marked increase in viscosity
even with the relatively small pressures obtainable in the laboratory.
THE LAWS OF ELASTICO-VISCOUS FLOW 407
The apparatus employed for the investigation consisted of
a light pulley with radius of 8 cm., over which passed two cords,
the ends of which carried scale pans for holding weights.
The specimen to be investigated had a diameter of 12 mm. at
the ends, while the intervening portion (75 mm. long) had a diame-
ter of 4mm. One end was clamped to the supporting frame and
the other to the pulley, which rests on a knife-edge in the axis.
The tests consisted in measuring the angular position of the
pulley by a micrometer at intervals of one minute while it is under
a constant torque.
LAWS OF ELASTICO-VISCOUS FLOW
The behavior of any solid under stress may be considered as
the resultant of four elements: (a) the elastic displacement, (0)
the elastico-viscous displacement, (c) the viscous displacement,
(d) the lost motion. These will be considered in turn.
a) The elastic displacement.—This is characterized by being
approximately proportional to the stress and independent of time."
A closer approximation is given by
iS: a C,PehP*
b) The elastico-viscous displacement.—This is manifested in a
_ slow return when the stress is removed; and it is assumed that the
same forces are brought into play during the direct motion.
« Doubtless there is some viscous resistance to this displacement, but it is very
small compared with that of cases b and c.
* The symbols used in this discussion are:
S =displacement (twist).
P=applied torque.
F=force, stress.
C=tfunctions of 6.
T =melting-point.
6=temperature.
t=time.
f)=duration of previous strain.
E, h, k, 4, Po, a, b, m, ™=constants.
e= Napierian base.
408 A. A. MICHELSON
This displacement is represented by the formula
Sa Ane —e-*V 1),
where
A,=C,Pe?,
‘c) The viscous displacement.—Here the elastic force is absent
or very small in comparison with the viscous resistance. The
specimen does not return to zero even after a long time interval.
The viscous displacement is given by
S3= (Fit Foto)? — (Foto)?,
in which F=C;Pe", and F.= the corresponding value, when P
has the value P, during the time 4.
For a specimen which has not been subjected to previous
strain the formula reduces to
S3= (Ft).
Experiment gives p=% approximately, until the specimen is near
the rupture point, when p approaches the value unity.
d) The lost motion.—If the stress be applied for a short time
(even a small fraction of a second), the specimen does not return
to the original zero. The difference between the original and the
new zero is the lost motion L.
It seems probable that the lost motion may be considered as
a function of ¢ such as #’, where 7 is very small (less than 0.02 for
zinc).
If this be considered as part of the viscous term
S; =A 3f(d),
then the total viscous yield may be represented by
S3=Ajl fl) +e]
(if the actual stress is between the limits o and P,, c=0).
«In some cases it may be made to return to the original position by heating or
by alternation (alternate positive and negative diminishing stresses).
THE LAWS OF ELASTICO-VISCOUS FLOW 409
THE RETURN
If after a time /, the displacement has reached the value S and
the stress is released, the specimen promptly returns to a displace-
ment short of zero and continues much more slowly in the same
direction.
If the elastico-viscous displacement at the time /, is given by
S,=A,(1—e72V *),
the corresponding return displacement at the time ¢, counted from
the instant of release, will be
R=A,e~*V *(1—e7*V *),
To account for the viscous term, assume
F=cS"S
s=|2 rat’ ae ae
<a pe LT wae
If F=constant, and F,=the constant value of F during the pre-
ceding stress during the time 4,
whence
Ss= gl (FH Foto)? — (Foto),
counting from the actual zero.
As shown by the formula, if the previous strain be considerable,
the new strain is relatively small. This strengthening by previous
strain is one of the striking features of the behavior of every sub-
stance which exhibits viscous yield.
If, in this expression, / represents the actual stress, it assumes
‘that the viscous force is proportional to the velocity, which is true
for fluids; but for “solid friction” the force is independent of the
velocity.
It may be assumed in the present case of internal viscosity of
solids that the actual law may be between these two extremes, e.g.,
P=a(S’)k,
* Experiment gives p=% (0.3-0.6), which makesw=1. The usual assumption,
n=o, gives p=1I.
410 A. A. MICHELSON
in which K<1. This would give P” instead of P, or, in better
agreement with experiment,
F’=CPe??.
The elastico-viscous term is readily obtained by making the vis-
cosity coefficient a function of the time.
Thus, if the restoring force be represented by aS, and the vis-
cous resistance by e#”S,* the integration gives
S=S.( 2 <7"),
a
where Or and r= —m+1.!
To determine the effect of temperature, the behavior of zinc,
glass, ebonite, pitch, and wax was studied. The results, together
with the preceding, may be summarized in the following formulae:
S=A,+A,7,+4;T3f
A,=C,PehP C,=E,+ Exe*®
A,=C,PeP C.= Eek
An CePesk C,—=Ff—E 0 r—e)x
T,=1—e7*V* h=b0
1 BON OND ee a
T,=C+ (+7) a (74) P=Pot
+(59)
* The assumptions in both viscous and elastico-viscous hypotheses make the
viscosity coefficient (that is, the coefficient of S) zero at the beginning of the motion
and infinite at =a, which is, of course, inadmissible. Instead of S” and t” we might
: B=S* Pao Peg 63 r :
substitute bse and ape which > and 7; are very small; but the resulting equa-
tions are far less simple and are not appreciably more accurate in expressing the results
of experiment than those here given.
+ The usual assumption, m=o, gives r=1I.
t Instead of this series coupling, the following may be substituted: The unit
consists of four elements: (1), (2), and (3) are in viscous contact with (4); (1) and (2)
are in elastic coupling; and, finally, (3) of this unit is connected with (1) of the next
following unit by an elastic coupling. The resulting formulae, however, are not
essentially different from those here given.
THE PHYLOGENY AND CLASSIFICATION OF
REPTILES
S. W. WILLISTON
University of Chicago
Not many years ago it was the fashion to construct phylogenetic
trees, often of wonderful design, for almost every group of animal
and vegetable life. Because of the failure of so many of them, the
practice has somewhat fallen into desuetude in recent years, and
it is only hesitatingly that I have ventured, for the first time, to
express in tabular form my own views of the phylogeny and classi-
fication of the reptiles. I do so the more readily, however, as I :
have no startling novelties to offer.
Phylogenetic schemes are always useful when constructed with
proper discrimination and with due regard to the known and the
unknown. They often furnish some residue of permanent knowl-
edge, some real contribution to taxonomy and the doctrine of
evolution; or, if not, by their failure they limit the field of legiti-
mate speculation. 7
Especially has our greatly increased knowledge of the early
land vertebrates, both at home and abroad, rendered it possible,
I believe, to approximate more closely the real origin of many
forms of vertebrate life than it was a dozen years ago. It was not
long ago that we were seeking the beginning, or at least the early
stages, of all reptile life in the order Rhynchocephalia. Because
we found, or thought that we found, in Sphenodon, or Hatteria, as
the genus was long called, the most generalized or primitive char-
acters among living reptiles, it was not unnaturally assumed that
its immediate ancestral stock was the most primitive reptilian
type of the past. And when Credner discovered a quarter of a
century ago, far back in the Permian rocks, another very primitive
reptile, it was also assumed, too readily, that it was of the same
stock. Upon that error, and it was an error, was built an elaborate
edifice with the Rhynchocephalia as its cornerstone, until, as so
4II
S. W. WILLISTON
412
UMOUNU = Q19yMaS| J \uooo0008
POO YON
pean Fears 77 uh ea Li
ne SIAN
S4.LISQVIG
- HUH
viShiaydomeg
Biyeydao
VIWISODA13d|,,
VIMNVSOLAHd
-VOISdv10
=~ BeanBsoyrsy
eimeSOUeBOIg”** -VOISdVUVd
VIMAVSORSTAUE
UBINOAaC
NOILNGIALSIG GNV ANSDOTAHd
lives 30
PHYLOGENY AND CLASSIFICATION OF REPTILES 413
- often in such cases, it was found that the foundation was insecure,
and the edifice toppled. Everything was made to fit the uncertain
base. Baur made the Proganosauria an integral part; Baur and
Osborn found in the ichthyosaurs a mere wing; Broom and Osborn
added a true cotylosaur; and Baur and Case built in the
pelycosaurs; Protorosaurus and Pleurosaurus were mere chinks;
and everybody (except Cope) united with them the lizards and
snakes. I do not mention these names in any invidious spirit;
they are all of men justly famous for their work in paleontology.
Fic. 2.—Sphenodon, skull from side and above. Diapsida. Recent
We all have the same proclivity, to find or to think that we find
those things for which we are seeking. As Baur himself has said,
they show “‘wie leicht man sich tatischen lassen kann, wenn man
durch eine allgemeine giiltige Anschauungsweise beeinflusst wird.”
A growing skepticism of the Rhynchocephalian affinities of Paleo-
hatteria disclosed little support for the far-reaching conclusions
based upon it; and, one by one, other assumptions have fallen
by the wayside. A careful examination of the type specimens of
Paleohatteria assured me that the genus was really a member of
the Theromorpha. Watson holds the same opinion; and Huene
has urged its relationship to the Pelycosauria.
It was Cope who, years ago, first suggested that in the temporal
region of the skull the surest criteria for the classification of the
Reptilia are to be found. Woodward carried the suggestion
ATA S. W. WILLISTON
' further, and showed their availability, but it was Osborn and
McGregor who first applied them definitely. They assumed too
much, as we have seen, but the credit is due to Osborn, more than
to anyone else, for the foundation of a true reptilian phylogeny,
and to him we owe especially a better knowledge of the double-
arched reptiles. He has called them the Diapsida, and there is
no better name for them. After the elimination of the forms which
we are sure do not belong with them, we are all now, I think, in
accord as to their phyletic unity. It is only in details that further
research (and there is much yet to be done) will be of value. The
separation of the great group called the dinosaurs, first proposed
by Seeley and warmly espoused of late by von Huene, into two
co-ordinate divisions, the Saurischia and Ornithischia, has much
to commend it. Valid arguments for the phyletic unity of the
pterosaurs, crocodiles, and pseudosuchians have been offered by
Huene, and their close relations with the phytosaurs and other
‘“‘thecodonts”’ is probable. All of these have certain annectant
characters, which, to me at least, are impressive. Altogether they
constitute the phyletic group of the Archosauria, a term we owe
to Cope.
The Diaptosauria, the other component group of the Diapsida,
is much smaller than when Osborn named it; and of the few forms
that are left, one, the Thalattosauria of Merriam, is discordant
and uncertain. It has been offered as a connecting link between
the true rhynchocephalians and the squamate reptiles; I would
rather shift it bodily to the Parapsida. The Diaptosauria are the
more generalized diapsids; the distinction between them and
the Theromorpha is not great. The earliest assured member of
the Diapsida has been carried back, I believe, no farther than late
Permian, in Youngina, to which Broom has given the inappropriate
group name of Eosuchia, inappropriate because Eosuchus Dollo
is a true crocodile.
The origin of the Diapsida, thanks chiefly to Baur and Case,
seems clear. These authors thought that the Pelycosauria were
really a part of the Rhynchocephalia, and for years they were
classed among them in our textbooks. It was an error, but the
error has shown, definitely I think, how the Diapsida arose, by the
PHYLOGENY AND CLASSIFICATION OF REPTILES 415
simple separation in reptiles with a lower temporal vacuity of
the orbito-squamosal arch, leaving an upper temporal opening.
The bones here are usually loosely connected, so loosely indeed in
Dimetrodon that it is only within the past few years that we have
become sure that the separation was not permanent; a permanent
separation, one that would admit a knife blade even, and the deed
is done. Whether or not Ophiacodon is a real example of this
beginning, and I believe that it is, from such forms as Mycterosaurus
the step to the diapsid type is trivial. If I am correct, and I am
confident that I am, we then have the origin of the double-arched
reptiles in early or middle Permian times from theromorph reptiles
with a single, typically lower, temporal opening, possibly from forms
not unlike Paleohatieria.
This group or subclass, which, with due modifications of the
original concept, may properly bear the name Synapsida given to it
by Osborn, includes scores of well-known genera of the orders
Theromorpha, Therapsida, and doubtless also the Sauropterygia.
It is the group that gave origin to the mammals, and has long
since been extinct. In its simplest and most primitive types,
together with all primitive characters of the skeleton, it has a
single temporal opening bounded below by the jugal, above by the
postorbital and squamosal. ‘This opening, I believe, arose by the
separation of the squamosal and jugal, and not by a definite
perforation of any bone. And this opening is the sole char-
acter by which the group is ultimately distinguished from the
Cotylosauria, its ancestral stock.
The evolution of the theromorph type through the dinoce-
phalian, therocephalian, and theriodont to the mammalian seems
assured by the African discoveries. In this evolution the structure
of the temporal region has undergone changes of which we do not
yet feel sure. The theriodont and anomodont reptiles, like the
plesiosaurian, may have the temporal opening extending from
the jugal to the parietal, apparently homologous with the combined
openings of the diapsid forms, or, as Watson has suggested, with
neither the synapsid nor diapsid; but, tracing the development as
a whole, I should sooner believe that they all arose from the prim-
itive type like that of Dimetrodon or Mycterosaurus, that is, from
416 S. W. WILLISTON
a typically lower opening bounded above by the postorbito-
squamosal arcade. In none of these forms does the quadratojugal
enter into the opening; in some it has entirely disappeared; and
the former, I believe, was the primitive condition. The opening,
I believe, first appeared below and behind the orbit, at the apex of
the squamosal. In the Cotylosauria the squamosal is a large bone
extending far down on the side of the skull and back of the quad-
rate. In Dimetrodon and Sphenacodon, indeed, the quadratojugal
is almost confined to the posterior side of the quadrate. Its
tendency was to disappear in this type of skull, and only in some
Diapsida did it become a part of the lower arch.
Fic. 3.—Mycterosaurus, skull, natural size. Synapsida, Permocarboniferous
‘
It seems now evident that the temporal openings arose in yet
another way: by the primitive separation of the postorbito-
squamosal arcade from the parietal in the stegocrotaphous skull;
and it seems very probable that this type of skull arose very early
in geological history, as early as the lower opening, and before the
separation of the upper arch in the diapsid skull. In these forms,
or in most of them at least, an additional temporal bone was
retained long after it was lost in other groups. And this is one
of the reasons why I believe that the Ichthyosauria and the Squa-
mata arose from a common or allied stem, direct from the Coty-
losauria. For this phylum I propose the name Parapsida. As
we have seen, Baur and others, because of the many primitive
characters of the Ichthyosauria, believed that the order came from
the original double-arched stem, that the lower temporal opening
had been secondarily closed. Cope, in 1896, asserted that the
PHYLOGENY AND CLASSIFICATION OF REPTILES 417
ichthyosaurs had an independent origin from the Cotylosauria, and
so indicated in a phylogenetic diagram. Broom adopted this
view in roor. In 1904 I quoted Cope’s views with approval.
More recently von Huene has reached the same conclusion, finding
in the Proganosauria or Mesosauria, as here accepted, either the
ancestral stock or one closely allied to it. Baur, it is true, in 1887
considered the Proganosauria as the ancestral stock of the ichthy-
osaurs, but Baur’s Proganosauria later included Paleohatteria,
believed to be a double-arched reptile. Merriam accepted the
Cotylosaurian ancestry, and Sollas very recently has voiced his
approval.t Indeed, the opinions now seem to be unanimous and
further discussion is superfluous.
Fic. 4.—Araeoscelis, skull, from side (A), and above (B), natural size. Parapsida
Permocarboniferous.
No order of reptiles has been the subject of more dispute than
the Squamata. The apparent absence of the quadratojugal, and
the presence of an additional bone in the upper temporal region,
together with the freely movable quadrate, have been explained
in various ways. At one time the arch articulating with the
proximal end of the quadrate was considered the lower one, and
the two bones the squamosal and quadratojugal. Again, the
t Sollas makes a rather curious error in saying that I was prepared to accept
the view, with certain reservations, of the direct descent of the ichthyosaurs from the
Stegocephalia. What I said was that the two bones in the temporal region of the
ichthyosaurs point to a direct origin from the stegocrotaphous reptiles.
418 S. W. WILLISTON
lower arch with the quadratojugal has been supposed to be lost,
or replaced by a ligament, and the upper bones called by various
names: supratemporal, squamosal, prosquamosal, etc. This lat-
ter view is the one now generally accepted, and, according to it,
the lizards have arisen from a primitively diapsid type which, by
the loss of the lower arch and the acquirement of streptostyly, has
become secondarily single-arched. Twelve years ago I ventured
the opinion that the two temporal bones of the squamate skull are
the tabular and squamosal, the former a bone unknown or unrecog-
nized in other reptiles since Triassic times. This view has nowhere
obtained approval except by Broom. For years past von Huene,
Broom, and I have repeatedly urged that the Lacertilia are a more
primitive type of reptiles than the Rhynchocephalia, and, years
ago, I ventured the prediction that the order would eventually
be discovered in the Permian. ‘The discovery of Araeoscelis in the
Permocarboniferous of Texas seems to fulfil that prediction.
Avraeoscelis has a single temporal opening bounded quite as in the
lizards, but with a fixed quadrate, the broad temporal region below
unperforated. The cervical ribs are single-headed and attached
to the centra. I am convinced that the Araeoscelis type of skull,
by the simple emargination of the lower border of the squamosal
and the consequent streptostyly, gave origin to the Lacertilia.
Watson has also shown that Pleurosaurus, a Jurassic genus which
has long been located among the Rhynchocephalia, likewise has a
single, upper temporal vacuity, with a fixed quadrate, but the
squamosal narrower. He believes that the genus was ancestrally
related to the Squamata, though he differs from me in the inter-
pretation of the bones of the temporal region, adopting the original
Baur view of the presence of squamosal and quadratojugal. This
group I have called the Protorosauria, believing that Seeley was
correct in his original interpretation of Protorosaurus.
Whatever may be the interpretation, one thing is evident:
even earlier than the origin of the upper vacuity in the Diapsida
a simple upper vacuity, as in the lizards, had developed, but with
the temporal region imperforate below, and this type persisted in
Pleurosaurus to late Jurassic. If the Squamata did not originate
in this way, then Araeoscelis and Pleurosaurus and probably
PHYLOGENY AND CLASSIFICATION OF REPTILES 419
Protorosaurus must represent an entirely new order of reptiles,
which very properly may be associated with the Ichthyosauria in
the division I call the Parapsida. It. might be urged that the
Diapsida originated from such a type by the development of a
lower vacuity after the upper one had been evolved. The argu-
ments against this view are too many and too potent; I need not
repeat them.
Admitting three chief groups of reptiles arising in late Penn-
_sylvanian or early Permian times, we have yet another, one which
by general consent is ancestral to all later amniota, the Coty-
losauria, and their direct descendants the Chelonia.
Fic. 5.—Pantylus, skull, from side, three-fourths natural size. Anapsida
Permocarboniferous. —
In this group the temporal region of the skull is wholly imper-
forate, and for the most part completely roofed over, a group which
very properly may be called the Anapsida. It was Baur who first
asserted that the turtles could not have originated from reptiles
with a perforated temporal region. Cope derived the order directly
from the Cotylosauria, through the Chelydosauria, an order based
upon a misapprehension. I approved and emphasized Baur’s
views in 1904. Case and Hay both hold the same opinion, and
only recently Watson has forcibly and convincingly presented the
claims of Eunotosaurus, from the Permian, as a real connecting
link between the two orders. And Broom and others are of the
same opinion. This unanimity of opinion renders further discus-
sion superfluous; Watson has presented the arguments.
We have, then, at least four main divisions or subclasses of the
class Reptilia, all beginning in Paleozoic times, and all represented
420 S.W. WILLISTON
by their direct descendants today; the birds, crocodiles, and
tuatara of the Diapsida; the mammals of the Synapsida; the
lizards and snakes of the Parapsida; and the turtles of the Anap-
sida. They suggest the following linear arrangement of the
known groups, the doubtful or poorly known ones, perhaps entitled
to ordinal rank, printed in italics:
Anapsida Diapsida
Cotylosauria Rhynchocephalia
Chelonia Rhynchosauria
Synapsida Thalattosauria
Theromorpha Choristodera
Therapsida Phytosauria
Sauropterygia Pseudosuchia
Placodontia Crocodilia
Parapsida Pterosauria
Ichthyosauria Dinosauria
Squamata “ Hosuchia”’
Protorosauria (Araeos-
celidia, Acrosauria)
I am aware that other general phylogenetic schemes of the
Reptilia have been proposed, especially by Boulenger and Goodrich,
but long years of study of the reptiles has convinced me that,
while all may have features worthy of consideration, the’ chief
reliance must be placed upon the skull structure, especially that
of the cranial and temporal regions. As von Huene and others
have urged, these parts are the most conservative, and least liable
to homoplastic duplication. Next to the skull, the ribs are con-
servative. In all the Archosauria the double-headed dorsal ribs
are attached to the diapophyses. In the Diaptosauria (except
Thalattosauria), the Synapsida (except the Sauropterygia), and
Anapsida the rib tubercle articulates with the arch, the capitulum
with the intercentral space, while in the Parapsida, so far as known,
the ribs are attached more or less exclusively to the centra. Other
characters originally proposed as distinctive between the Diapsida
and the Synapsida in the wider sense have been proven to be invalid.
We know that the primitive foot structure is nearly that of the
lizards and Sphenodon of today, that the reduction of the phalanges
in the theriodonts and turtles is purely homoplastic. The supposed
relationship between the turtles and the plesiosaurs is also purely
PHYLOGENY AND CLASSIFICATION OF REPTILES 421
homoplastic; nor am I aware of any constant character in the
girdles, limbs, or ventral ribs. Goodrich, in a recent paper on
the phylogeny of reptiles, relies greatly upon the structure of the
feet, especially of the fifth metatarsal. I cannot accept his con-
tentions, some reasons for which I have published elsewhere.
Just when the animals we call reptiles arose in geological history
we do not know; certainly it was in early Pennsylvanian times,
probably in Mississippian. That they arose from what we call
the Amphibia, forms with temnospondylous vertebrae, is certain,
though there is probably not much more reason for calling the
ancestral stock Amphibia than Reptilia. I prefer to call it,
provisionally, Protopoda. It was ancestral to both, and both
classes have advanced since their divergence, the Amphibia some,
the Reptilia much. Could we find, as some time we hope that we
may, in mid-Mississippian or late Devonian times, a skeleton of
one of those ancestral creatures, we should perhaps not call it by
the name of any known order; it would be the old question over
again of the differences between animals and plants. At present
we know the Protopoda only by their footprints. As it is, we are
dealing chiefly with archaic forms, even by the close of Penn-
sylvanian times, forms which have retained in various degrees
their primitive characters while adding or losing others in different
ways. Just as the most primitive mammals now living have
become highly specialized in the loss of teeth, so too the amphibians,
as we know them in Paleozoic times, were more or less specialized
or degenerated. The only known distinctive characters between
the two classes, as represented by their known skeletons in Permo-
carboniferous times, are found in the atlas and feet, and doubtless
some time we shall find these differences bridged over. For a
long time we relied upon the open palate of the Amphibia, but
Watson has deprived us of that support, and now we are compelled
to group the characters at length in any differential diagnosis of
the two classes. However, with nearly every character in common
to the Amphibians and reptiles, we are never in doubt as to which
class a given form belongs if we know it well enough, for all the
common characters have never been found in the same specimen,
and doubtless never will be in any specimen from rocks later than
the Mississippian.
OUR PRESENT KNOWLEDGE OF ISOSTASY FROM
GEODETIC EVIDENCE
WILLIAM BOWIE
Chief, Division of Geodesy, United States Coast and Geodetic Survey,
Washington, D.C.
For a number of years investigations have been made by the
Coast and Geodetic Survey upon the subject of isostasy. The first
work was done under the direction of Professor John F. Hayford
in connection with a study of the deflection of the vertical and the
determination of the shape and size of the earth. The later work
consisted of investigations of the effect of topography and isostatic
compensation upon the intensity of gravity. This work was
started by Professor Hayford, and the first report on it was made
to the International Geodetic Association.’ The first comprehen-
sive report on this work was made by Professor Hayford and the
writer in Special Publication No. to of the Coast and Geodetic
Survey which appeared in 1912.2 The investigations of gravity
and isostasy were continued under the direction of the writer, and
the results have been published in two reports, one appearing in
1912 and the other in 1917.3
As is well known, the theory of isostasy postulates that at some
depth below sea-level forces are in equilibrium and, therefore, that
each column of unit cross-section extending from the depth of
compensation to the surface of the earth contains the same amount
of matter; or, to be more exact, it may be stated that each column
© Geodetic Operations in the United States, 1906-9, a report to the sixteenth general
conference of the International Geodetic Association. Separate publication of the
Coast and Geodetic Survey (not numbered), 1909.
2 Effect of Topography and Isostatic Compensation upon the Intensity of Gravity,
Special Publication No. to, Coast and Geodetic Survey, 1912.
3 Effect of Topography and Isostatic Compensation upon the Intensity of Gravity,
Second Paper, Special Publication No. 12, Coast and Geodetic Survey, 1912; Imvesti-
gations of Gravity and Isostasy, Special Publication No. 40, Coast and Geodetic Survey,
IQt7.
422
OUR PRESENT KNOWLEDGE OF ISOSTASY 423
of unit cross-section which extends from the depth of compensation
to the sea-level surface has the same weight.
One of the main objects of the investigations made in the Coast
and Geodetic Survey is to determine to what extent isostasy is
proved to exist. Other objects are to determine, if possible, the
method of distribution of the compensation, horizontally and
vertically, with respect to each topographic feature, and to discover,
if possible, the cause or causes of the gravity anomaly which cannot
be accounted for by the topography and by the isostatic compen-
sation of the topography. By topography is meant the material
above sea-level on the continents and islands and the deficiency of
density in the matter between the ocean surface and the bottom of
_ the oceans.
Professor Hayford made the following statement in his second
publication on the figure of the earth and isostasy. ‘‘One may
properly characterize the isostatic compensation as departing, on
an average, less than one-tenth from completeness or perfection.
The average elevation of the United States above sea-level being
about 2,500 feet, this average departure of less than one-tenth
part from complete compensation corresponds to excesses or
deficiencies of mass represented by a stratum only 250 feet thick,
on an average.’
Professor Hayford based his conclusion upon the fact that the
mean residual or deflection of the vertical, after the isostatic correc-
tion was applied, was 2.91 seconds of arc. After only the topo-
graphic deflection was applied, the residual was 30.37 seconds.
It is seen, therefore, that the application of the effect of the isostatic
compensation reduced the average residual from 30.37 to 2.91
seconds.
In making the corrections for the effect of the isostatic com-
pensation, Professor Hayford assumed that the compensation was
directly under the topographic feature and that it was distributed
uniformly to the depth of compensation. This depth of compensa-
tion was an unknown quantity, to be determined from the available
data. The depth derived by him is 122 km.
t Supplementary Investigation in 1909 of the Figure of the Earth and Isostasy,
special publication (not numbered) of the Coast and Geodetic Survey.
424 WILLIAM BOWIE
Professor Hayford stated that the anomalies or deflections of
the vertical, resulting after the application of the correction for
isostatic compensation as well as for the topography, would be an
indication of the extent to which the conditions postulated would
not be true.
Professor Hayford should have stated that the compensation
departed from perfection only 10 per cent locally, for there is no
indication that there is a departure of 10 per cent from the perfect
state for the whole country. Some of the anomalies were positive
and others negative and these tend to balance for the whole area.
Several tests were made by Professor Hayford to show the
result of other methods of vertical distribution of the isostatic
compensation than the one of uniformity. If the isostatic compen-
sation is uniformly distributed through a stratum 10 miles thick,
he found the most probable depth for the bottom of this stratum
to be 40 miles. If the isostatic compensation is distributed with
respect to depth, according to the law postulated by Professor T. C.
Chamberlin, the most probable value of the limiting depth is 193
miles. The method of distribution of the compensation by Pro-
fessor Chamberlin’s method is given on pages 159 and 160 of The
Figure of the Earth and Isostasy from Measurements in the United
States. This is the first report’ on the deflection of the vertical and
isostasy. In regard to the Chamberlin method, Hayford said: ‘‘It
is not possible to ascertain whether this compensation is more
probable than the solution G compensation, uniformly distributed
from the surface to a depth of 70.67 miles, since the two sets of
computed deflections agree so closely that their differences are much
smaller than the accidental errors.”
When the investigations of gravity and isostasy were under-
taken, it was concluded that the compensation should be distrib-
uted uniformly to a depth of 113.7 km., which was the depth
determined by Hayford in his first investigation of the deflection
of the vertical and isostasy. The uniform distribution was adopted
because it made easier the preparation of the tables with which the
computation of the effect of the isostatic compensation was made.
It was also believed that this method of distribution was as probable
Special report (not numbered) of the Coast and Geodetic Survey, 1909.
OUR PRESENT KNOWLEDGE OF ISOSTASY 425
as any other simple method of distributing the isostatic compensa-
tion. It was realized by Professor Hayford and by the writer of
this paper that uniform distribution to the depth of compensation
could not be true at all places; yet it was thought that the depar-
tures from normal densities in the lithosphere due to the compensa-
tion were heterogeneous and that for a large area the effect would
therefore be practically that of uniform deviation from normal.
What follows in this paper is largely based upon the results of
the most recent investigation of gravity and isostasy (Special
Publication No. 40). ,
In this investigation there were used 219 stations in the United
, States, 42 in Canada, 73 in India, and 4o other stations, principally
in Europe—374 in all.
There were certain phases of the investigation which had to be
confined to stations in the United States, owing to lack of data for
the other stations. These consisted of tests to show whether local
or regional distribution of compensation horizontally is the more
probable, and also of tests to derive the most probable value of
the depth of compensation based upon uniform distribution of the
compensation vertically from the surface of the ground to the
depth of compensation.
Three hundred and fifty-eight stations were used for the deter-
mination of the most probable values for the constants in the gravity
formula which gives the value of gravity, y,, at sea-level for any
latitude, @. This formula is
Yo= 978.039 (1+0. 005294 sin? 6—0.000007 sin? 2 ¢).
It is a remarkable fact that this formula agrees almost exactly with
the formula derived in 1912 from only 122 stations in the United
States alone. That formula was
Yo=978.038 (1+0. 005304 sin? 6—0. 000007 sin? 2 ¢).
This agreement between the derived formulas is a clear indication
that isostasy is present to practically the same degree in | One
countries as it is in the United States.
426 WILLIAM BOWIE
The average gravity anomaly with regard to sign for stations
in the United States by the adopted method of reducing for isostasy
is —o.003 dyne, while the average without regard to sign for these
stations is 0.020 dyne. An anomaly is the difference between the
observed and computed values of gravity. The computed value
has corrections applied for the elevation of station above sea-level
and for the effect of the attraction of the topography of the whole
world and of the opposite effect of the isostatic compensation of
the topography. The topography is considered to be that material
on the continents and on islands which is above sea-level and the
deficiency of material in the oceans.
An anomaly of 0.020 dyne in terms of mass is here given in
order that the reader may have a clear conception of the magnitude
of the deviation of the gravity from normal. If we should have a
disk of material directly under a gravity station and if the disk
should be of normal density, 20 km. in diameter and about 600 feet
thick, the attraction on a gram mass at the station would be 0.020
dyne. An anomaly of 0.001 dyne represents the attraction of a
disk of material of indefinite horizontal extent and about 30
feet in thickness on a gram mass located near the center of the
surface. :
The anomalies by the isostatic method varied from +-0.059
at Minneapolis, Minnesota, to —o.093 dyne at Seattle, Washington.
There were only ten anomalies which were greater than 0.050 dyne.
When a correction was applied to the computed value of gravity
for the effect of the topography, but none for the isostatic com-
pensation, the mean anomaly with regard to sign for the stations
in the United States was —0.037 dyne. The corresponding mean
without regard to sign for the stations in the United States was
0.050 dyne.
The fact that the gravity anomalies were more nearly eliminated
by the isostatic method of reduction than by the methods where
isostatic compensation was not considered is strong evidence in
favor of the former method.
We must conclude, however, that no method can be near the
truth unless it has a rather general application to different sections
of the country and to different classes of topography. In order
OUR PRESENT KNOWLEDGE OF ISOSTASY 427
to test the methods of reductions by this theory, the stations were
arranged in five groups according to the topography. The anom-
alies with and without regard to sign for the several classes of topog-
raphy and for the two methods of reduction—one taking into
account the topography and compensation, and the other only
the topography—are given in Table I. To make it easier to refer
to these methods the first will be called the Hayford method and
the other the Bouguer method.
TABLE I
MEAN ANOMALIES
MEAN ANOMALIES
With Regard to Sign | Wi i
CHARACTER OF STATIONS NUMBER QE ; ee ui: oc aoe
Hayford; Hayford;
Depth, Bouguer Depth, Bouguer
113.7 km. 113.7 km.
CoastistationSe adie seen merce 27 —0.009 | +0.017 | 0.018 0.021
Stationsimear coast. | aas,..: 40 — .oor | + .004 .O21 .025
Stations in interior, not in
mountainous regions....... 88 — .001 ; — .028 .O19 .033
Stations in mountainous re-
gions, below general level. . . 36 — .003 | — .107 .020 .108
Stations in mountainous re-
gions, above general level... 20 + .00or1 | — .110 .O17 .III
All stations (except the two
Seattle statioms)/ ene crnis. 217, 1) 002i) 026 .O19 049
PAN sta tlonse mastseye st ayer eben: 219 —0.003 | —0.037 | 0.020 0.050
This table shows that the Hayford reduction gives about the
same values without regard to sign for each class of topography
while the mean anomalies with regard to sign have a very small
range if we do not consider the 27 stations at the coasts. It is prob-
able that the coast stations are affected by the presence of Cenozoic
formation, the material of which is lighter than normal. This will
be referred to later. We may conclude, I think, that there is prac-
tically no relation between the sign and the size of the Hayford
anomalies and the character of the topography on which the stations
are located.
We find entirely different conditions in regard to the Bouguer
anomalies. The size of the anomaly without regard to sign at
428 WILLIAM BOWIE
coast stations is 0.021, practically the same as for the Hayford
method of reduction. ‘This is, of course, due to the fact that there
is very little relief in the topography at the coast. The size of the
Bouguer anomaly for stations in the mountainous regions above
the general level iso.111 dyne. The range is therefore 0.090 dyne.
When we consider the mean anomaly with regard to sign for the
Bouguer method of reduction, we find a range for the groups from
+o.017 to —o.110 dyne. This is a total range for the groups of
0.127 dyne. If we should consider the individual stations, we
should find much wider ranges for the Bouguer values than for the
Hayford values. The total range for the Hayford values is from
+o.059 to —0.093. ‘The total range for the individual stations for
the Bouguer reduction is from +0.057 to —o. 229.
The values given above are conclusive proof that the condition
of isostasy exists to a rather remarkable degree, and that the theory
that the topography of the earth is not compensated for by a lack
of density under the continents and by an excess of density under
the oceans is far from the truth.
The fact that the country as a whole is in a high state of isostatic
adjustment is evident from the values given above, but there are
local deviations from normal which may be due to a number of
causes. ‘They may be due to departures from the state of perfect
isostasy, to an erroneous method of distributing the compensation
horizontally from the station, to an erroneous method of distribut-
ing the compensation vertically with respect to depth, to erroneous
values employed for the density of the topography, or to an
erroneous depth of compensation; or they may be due to the pres-
ence of material heavier or lighter than normal close to the station
but below sea-level. This extra or deficient density may or may
not be compensated for in lower portions of the lithosphere.
There were made, during the investigation, certain tests which
throw some light upon the causes of the gravity anomalies.
It has been held by some that the compensation of topography
is not distributed locally under the topographic feature, but is
extended horizontally to some unknown distance. It does seem
improbable that the compensation should be directly under the
topographic feature and not extended horizontally to a certain
OUR PRESENT KNOWLEDGE OF ISOSTASY 429
extent. At the same time, it is equally improbable that the com-
pensation should be extended horizontally and uniformly out to any
definite distance froma station. It would seem to be more probable
that the compensation is distributed regionally, with the greater
amount of the compensation directly under the topographic
feature, and that it diminishes in amount with the distance from
the feature.
A test was made to show whether local distribution or regional
distribution was the more probable. ‘The distribution in each case
in the regional method was uniform. ‘The method employed was
to take the average elevation of the topography within a certain
distance of the station. In one case the distance was 18.8 km., in
another 58.8, ina third case 166.7 km. from the station. With the
average elevation within these areas, a computation was made of
the effect of the compensation, which was supposed to be uni-
formly distributed out to the limit of the area and also uniformly
distributed from the surface to the depth of compensation.
The method of distributing the compensation horizontally
necessarily leads to some error, for, as a matter of fact, the com-
pensation of each topographic feature should be distributed
regionally with respect to that particular feature. But such com-
putations would be extremely laborious and it is believed that the
results would not be materially different from those which were
obtained. ‘That erroneous results might be obtained for a single
station is readily perceived when we consider that the station may
be on a plain or plateau, say 167 km. in radius, and that just
outside of this area there are massive mountain masses. According
to the theory of regional distribution of the compensation, the
compensation of the mountain masses should be extended under the
plains; it should therefore have an effect on the computed gravity at
the station. It would tend to make the computed value of gravity
at the station smaller than it would otherwise be. On the other
hand, we might have a station in a mountain mass of, say, 167 km.
in radius, with plains surrounding the mass. In this case all the
compensation of the mountain mass would be used in the computa-
tion of the corrections to gravity. Some of the compensation
should be distributed for a distance of 167 km. out into the plains.
430 WILLIAM BOWIE
There were 124 stations in the United States which were used
in the investigation of the regional distribution of compensation.
We shall speak of the three regional distributions as Zones K, M,
and O,* since the distances given above are the outer limits of those
zones. The anomalies for the local and for the three regional
distributions of compensation are shown in Table II for the 124
stations.
The values in this table give no evidence whatever in favor of
any one method of distribution over the others. This is probably
as it should be, for most of the stations considered were in topog-
raphy of low relief. When a station is on a plateau or a plain, it
TABLE II
ANOMALIES
Regional
Local
Zone K Zone M Zone O
Mean with regard to sign.......... —0.002 —0.001 —0.001 —0.002
Mean without regard to sign....... ©.020 0.019 0.020 0.020
is evident that the method of distribution of the compensation has
very little effect on the anomaly. If the topography were of exactly
the same elevation throughout the zone, the local and regional dis-
tribution of compensation would give absolutely the same value.
The difference in the effect would increase with the increase in the
difference in elevation of the topography in different parts of the zone.
While some of the stations might have larger anomalies by some
one of the methods than by the others, there would be other stations
for which the reverse would be true, and the mean anomaly for all
the stations by each method would necessarily tend to be the same.
We may assume, as was done when considering the relation of
the anomaly to the topography, that that method of distribution
is most nearly the truth which has the smallest variation in anom-
aly for different classes of topography.
t This refers to the zones used in computing the effect of topography and com-
pensation. See Special Publication No. to.
OUR PRESENT KNOWLEDGE OF ISOSTASY 431
The 124 stations under consideration were arranged in five
groups, depending upon the topography. ‘The anomalies with and
without regard to sign for the local and the three regional distribu-
tions of the compensation are shown in Table ITI.
The mean anomaly with and without regard to sign for the
several groups and for the various methods of distributing the
TABLE III
LocaL AND REGIONAL ANOMALIES
Locat Com-| ReEGIonAL CoMPENSATION ANOMALIES
PENSATION
ANOMA-
LIES Zone K Zone M Zone O
FOR 18 COAST STATIONS
Mean with regard to sign.......... —0.004 | —0.004 —0.004 —0.006
Mean without regard to sign....... 0.018 0.018 0.018 0.020
FOR 25 STATIONS NEAR THE COAST
Mean with regard to sign.......... —0.002 —0.001 —0.001 —0.001
Mean without regard to sign....... 0.022 0.021 0.021 0.022
FOR 39 STATIONS IN THE INTERIOR, NOT IN MOUNTAINOUS REGIONS
Mean with regard to sign...
Mean without regard to sign
co ee ee
+o.002
0.018
+o.oo1 |
0.017
+o0.002
0.018
+o.003
0.017
FOR 22 STATIONS, IN MOUNTAINOUS REGIONS, BELOW THE GENERAL LEVEL
Mean with regard to sign...
Mean without regard to sign
+o.oo1
0.017
©.000
0.017
+0.006
0.019
+0.003
0.018
FOR 18 STATIONS, IN MOUNTAINOUS REGIONS, ABOVE THE GENERAL LEVEL
Mean with regard to sign..........
Mean without regard to sign.......
0.000
0.017
+0.003
0.018
=O .OLO
0.020
compensation does not throw very much light upon the question
of which method is nearer the truth, although a careful analysis
of the table will, it is believed, indicate that the distribution
regionally to the outer limits of Zone O, 166.7 km., is not as
probable as the local distribution of compensation or the regional
distribution out to the limits of Zone K, 18.8km., or Zone M,
<
432 WILLIAM BOWIE
58.8km. The greatest range for the mean anomalies with regard
to sign is for Zone O and for mountainous regions where the
stations are below and above the general level. The range here is
from -++-o.006 to —o.o10, which is 0.016 in all. This range is
almost as large as the average anomaly for the United States
without regard to sign. The largest range for the other methods
of distribution is 0.007.
We should expect the regional and local anomalies to be approxi-
mately the same for all of the stations not in mountainous regions,
but we should also expect that the local and regional anomalies
would differ for the stations where the relief is great. Where the
station is below the general level, the computed value of gravity
is less because some of the compensation of the mountains is dis-
tributed closer to the station than it would be by the local distribu-
tion. Where the station is above the general level, the computed
value of gravity is greater because some of the compensation of the
local topography is distributed under the contiguous valleys.
We may conclude, I think, that the solution of the problem of
local distribution of compensation of the topography is indeter-
minate to a certain degree; that is, that any one distribution is as’
probable as any other one, out to a distance of 59 km. from the
station. It is possible and, in fact, probable that this uncertainty
may extend to a distance somewhat greater than 59 km., but it is
very probable that it does not extend to a distance of 167 km. from
the station. This conclusion is based upon the geodetic evidence,
as furnished by the gravity anomalies, and has no connection with
geological evidence. Any decision as to whether one method or
another is the more probable within the distance of about 60 km.
should be left to the judgment of geologists. It is of course possible
that, with more geodetic data available, geodesists may be able to
throw additional light on this subject.
It should be remembered that when ales the test for the
most probable method of distributing the compensation horizontally
the anomalies are treated as if they were due only to the method of
distribution. Asa matter of fact, they are probably due to a num-
ber of causes, and this fact has some effect on the results of the com-
putation, but it is believed that its effect is small.
OUR PRESENT KNOWLEDGE OF ISOSTASY 433
When the researches in isostasy and in the deflection of the
vertical were started, it was assumed that the compensation of
topography was complete, that it was distributed directly under the
topographic feature, and that it was uniform with respect to depth.
This method of distribution vertically and horizontally was contin-
ued in the computations connected with the researches in gravity
and isostasy. The depth of compensation derived from investiga-
tions of the deflection of the vertical, which was considered to be
of the greatest probability, was 122.2 km."
As it is probably true that the gravity anomalies may be due to
an erroneous depth of compensation, it was decided to compute a
new value for the depth, based upon gravity observations alone.
The results of these computations, with numerous tables, are given
TABLE IV
DEPTH OF COMPENSATION IN KILOMETERS
42.6 56.9 85.3 TEQe7, 127.9 |. 156.25 184.6
Mean anomaly with re-
gard to sign, for all
StATIONS Merce acto +0.004/+0.003} 0.000/—0.003}/—0.004|—0.007|/—0.010
in Special Publication No. 40, and only the results need be given in
this paper. In the investigations from which were determined the
most probable depths from gravity determinations, there were
used only the 219 stations in the United States. It was impossible
to use the gravity stations outside of the United States because
of lack of detailed data.
The depth 113.7 km., the one derived in the first investigation
of The Figure of the Earth and Isostasy from Measurements in the
United States, was used in making the reductions for compensation
in the investigations of gravity. With the detailed information
obtained from the computations it was possible to obtain the effect
of the compensation for other depths. This was done by means of
certain factors.
The mean anomaly with regard to sign for all the stations is.
shown in Table IV. The means without regard to sign for all of
tSee Supplementary Investigation in 1909 of the Figure of the Earth and Isostasy,
p. 54. ;
434 WILLIAM BOWIE
the stations used as a single group for the various depths of com-
pensation had a range of only 0.002 dyne from 0.020 to 0.022.
Owing to the small range, it is not necessary to show the means
without regard to sign in the table. The anomalies in this table
have not a very wide range, and therefore there is no very strong
evidence to show that any one depth among the intermediate depths
is much better than the others. The evidence, such as it is, favors
the depth of about 85 km.
TABLE V
ANOMALIES AT DIFFERENT DEPTHS OF COMPENSATION
Mean Anomalies with Regard to Sign for Depth in Kilometers
Character of Numba
Topography, | “1.5.8” |
42.6 56.9 85.3 DIZe7, 127.9 156.2 184.6
Coaster eles 27 |—0.002}—0.003|—0.006!—0.007|—0.008]—0.009|—0.009
Near coast...... 46 |+ .002/+ .002/+ .oo1\/+ .oorj/+ .oo1/-+ .oot/-+ .oor
Interior, not in
mountainous re-
PIONS eee ee 87 |— .003/— .002/— .oo1/+ .oo1|-+ .oor|-+ .003/-- .005
Mountainous _ re-
gions, below -
general level...| 36 .000 .000 .000/— .001]/— .oo0I]/— .002/— .0a3
Mountainous _ re-
gions, above
general level...) 20 |+-0.021|/-+0.016/-++0.009|-+0.003|}-++0.001|—0.003} —0.006
In order to get stronger evidence as to the most probable depth
of compensation, the stations were divided into groups according
to the topography, as was done in several other tests. The mean
anomalies without regard to sign for the seven depths considered
had a total range of 0.009 from 0.017 to 0.026. Each of these
mean anomalies came in the group of stations in mountainous
regions, above the general level. The maximum range for any other
group of stations was 0.004 and that occurred in the mountainous
regions where the stations are below the general level.
The means with regard to sign varied for the different depths
and the different classes of topography. These means are shown
in Table V.
It will be noticed that there is only one group for which there
are very decided changes in the mean anomaly with regard to sign—
OUR PRESENT KNOWLEDGE OF ISOSTASY 435
stations in mountainous regions, above the general level. The
anomaly is +o.021 for a depth of 42.6 km. and it is —o.006 for a
depth of 184.6 km.
Computations were made to obtain the most probable depth
from all the gravity data for the 219 stations in the United States.
Where all stations were used, the depth was found to be 67.1 km.
It was realized that the stations on topography which was not in
mountainous regions were not well adapted for the determination
of a depth of compensation. Owing to the low character of the
topography, the effect of the compensation was nearly the same
regardless of the distance from the station to which it was extended.
This is due to the fact that the attraction of an indefinitely extended
disk containing a certain mass but of indefinite thickness will exert
the same attractive force on a given mass regardless of how far
from the disk the mass is placed. It is assumed, of course, that the
attracted mass is over the center of the disk. We can see, therefore,
that where the compensation is nearly the same in amount under a
unit area for an indefinite distance around the station it would,
although a number of kilometers in thickness, attract the pendulums
at a station by the same amount regardless of the thickness of the
disk or column of compensation.
The condition is different in the mountain regions for those
stations which are above the general level, for there the topography
near the station is comparatively limited in horizontal extent and
the attraction of the compensation will depend upon the depth to
which the compensation is extended. ‘The closer to the station the
greater of course will be the effect of the compensation and the
larger will be the plus value of the anomaly observed, minus com-
puted gravity. Where the compensation is extended to a great
depth, the effect of the compensation is decreased, the computed
value of gravity is necessarily larger, and the anomaly tends to be
negative.
After consideration of all the facts, it was decided to determine
the most probable value of the depth of compensation from gravity
data by using only the stations in the mountainous regions below
and above the general level. This was done and the depth resulting
was 95 km.
436 WILLIAM BOWIE
It is interesting to note that, in the deflection of the vertical
investigations, depths of compensation were determined for a num-
ber of groups, in addition to a depth for the whole country using
all the stations as one group.’ When we use the groups which are
in mountainous regions and give the value of the depth of compensa-
tion derived from each group the same weight, it is found that the
depth of compensation, as derived from deflections of the vertical
data, for stations in mountainous regions only, is 97 km. It is
rather remarkable that practically the same depth should be ob-
tained from such widely different geodetic data. It is believed
that the mean of these two values, or 96 km., is about the best
value that is available at present from all geodetic data. As in
other tests made during the investigation of gravity and isostasy,
it was necessary to assume that all of the anomalies were due to the
erroneous depth of compensation when the derivation of the most
probable depth was made. This necessarily places some uncer-
tainty in the depth of compensation, although it is believed that
the uncertainty due to that cause is moderate in amount. It is
probable that the best depth of compensation which will be derived,
from more geodetic data will be somewhere between 80 and 130 km.
This, of course, is on the theory that the compensation is distributed
uniformly from the surface or from sea-level to the depth of com-
pensation.
The writer does not believe, as was stated earlier in this paper,
that the compensation is distributed locally and uniformly to the
depth of compensation. It is possible that the compensation may
be distributed by some other method. It is probable that there
is no method of distribution that is general, that is, applicable to
each local area in the country. It seems to be most probable that
the compensation varies from place to place and that the greater
portion of the compensation may be near the surface in one place
and lower down in another, or that it may be distributed through-
out a considerable depth with varying amounts at different depths.
A computation was made, but the results of this do not appear
in Special Publication No. 40, which showed the depth of the disk
t See Supplementary Investigations in 1909 of the Figure of the Earth and I sostasy,
p. 58.
OUR PRESENT KNOWLEDGE OF ISOSTASY 437
within which all of the compensation should be concentrated in order
to have its attractive effect equal to the effect of the compensation
uniformly distributed from the surface of the earth to a depth of
113.7km. In other words, if all of the compensation were con-
tracted to the disk at the particular depth, it would have the same
effect as the uniform distribution.
If the compensation is distributed regionally to a distance of
10 km. from a station the disk within which all of the compensation
is supposed to be concentrated must be placed 21.3 km. below the
station. With regional compensation distributed to a distance of
20 km., 60 km., or too km. from the station, the depth of the disk
becomes respectively 28.6 km., 41.2 km., or 45.5 km. If the com-
pensation is started at sea-level instead of at the surface of the
ground, each of the depths given above should be increased by
about 1 km.
These depths are of particular significance, for they represent
what may be called the effective center of the compensation on the
basis of uniform distribution with respect to depth and with a
depth of 113.7km. This depth, as shown by certain tests, gives
practically as good results as what may be called the most probable
depth of 96km. It is significant that there can be a variation in
the depth of as much as 18 km. without materially affecting the
anomalies.
It is reasonably certain that the effective depth as given above
would be practically the same for all of the intermediate depths
used in the computations to show which was the most probable
depth. We may conclude, therefore, that the figures given above
actually represent the effective center of the compensation, regard-
less of the method of distribution of the compensation. If, for
instance, the compensation were considered to be confined to a
zone about 20 km. in thickness, the center of that zone would have
to be between 30 and 50 km. below sea-level. If the compensation
is distributed according to the Chamberlin method,’ the greater
portion of the compensation would necessarily have to come within
100 km. of the surface, but there would be part of it at some
distance below that depth.
tSee The Figure of the Earth and Isostasy from Measurements in the United States,
p. 160.
438 WILLIAM BOWIE
It would not be a very difficult matter to draw curves represent-
ing different methods of distribution of compensation which would
have effective depths of the compensation equivalent to those shown
above for uniform distribution. If we have the effective center of
compensation about what it is for the uniform distribution, then,
under any method of distribution of compensation, the greater
portion of the compensation would be between the sea-level surface
and about too km.
It may be concluded from a study of the gravity data and also
of the deflection of the vertical data that there is no geodetic evi-
dence which favors any particular method of vertical distribution of
compensation. Anyone is therefore free to use a method of dis-
tribution which best serves his purpose or which may fit the par-
ticular theory he may hold in regard to the constitution of the
earth’s lithosphere. But, in order to secure results which are as
accordant as those given in the latest report of the Survey, the effec-
tive depth of compensation must be between 30 and 50 km.
It was noticed early in the investigations of gravity and isostasy
that there were apparently some relations between the gravity
anomalies and the densities of the materials at the surface of the
earth close to the station. This subject was treated briefly in
Special Publications Nos. 10 and 12, which gave the results of the
earlier investigations of gravity and isostasy. With the additional
material available from other countries as well as from the United
States for the most recent investigations of gravity and isostasy,
these relations between the gravity anomaly and the surface density
are shown to be stronger.
In the United States the stations on the dense rock which
belongs to the pre-Cambrian formation have anomalies which tend
strongly to be positive. This is an indication that under the station
the material of this formation extends to a considerable depth where
the gravity anomaly is large.
It was found that the stations on the pre-Cambrian formation
in Canada did not have the tendency to be positive that was shown
in the United States. This may be due to the extensive areas
covered by this formation in Canada. As was stated above, the
attraction of a disk of material of indefinite extent is independent
OUR PRESENT KNOWLEDGE OF ISOSTASY 439
of the distance of the attractive mass from the disk; therefore, if
we should have in Canada a station on an extensive area of pre-
Cambrian formation where the material is of uniform thickness,
and if this material were compensated for by a deficiency of material
below it, then the compensation would have an effect which
would practically counterbalance the effect of this denser material
which is near the surface.
In India there are only eight stations on the pre-Cambrian
formation, six of them having positive anomalies and two negative.
But the mean with regard to sign of the anomalies is nearly zero.
It may be possible that the small number of stations on this forma-
tion in India prevents the stations there from showing the same
relation to the formation that we have in the United States. It is
worthy of note that each of the areas of the pre-Cambrian forma-
tion in the United States on which stations are located is rather
small in horizontal extent. If we should have a pre-Cambrian
formation 10,000 feet thick under a station with a density of the
rock to per cent above normal, and if the formation extended
10 km. in all directions from the station, the effect of the increased
density would be to increase gravity by +-o.029 dyne. If this
extra material were completely compensated for and the compen-
sation were distributed uniformly to a depth of about 114 km.,
the negative effect of the compensation would be —o.003 dyne.
The resultant would be +0.026 dyne, which is about the size
of the average pre-Cambrian anomaly in the United States.
It was found that the anomalies at stations on the Cenozoic
formation had a tendency to be negative both in the United States
and in India. There were only two stations in Canada on this
formation and they were both negative. It is probable that the
reasoning employed above in regard to the pre-Cambrian anomalies
will apply to the Cenozoic anomalies. The density of the material
of this formation is in general about 5-10 per cent less than normal,
and the presence of this light material near the station should have
a greater effect dn the value of gravity than the compensation of
this material, if any, which would be lower down in the lithosphere.
If the Cenozoic formation should be of great horizontal extent
and of uniform thickness, the effect of material of this formation
440 WILLIAM BOWIE
would be offset almost entirely by the effect of the compensation
if this lighter material were compensated for by an excess of
material lower down in the lithosphere.
The mean anomaly with regard to sign for the Cenozoic stations
in the United States was —0.007 dyne. In India it was —o.017
dyne. There were 31 stations in India on this formation and 20
had negative anomalies and 11 had positive ones. The positive
anomaly in every case was comparatively small, the largest being
0.033 dyne. There were to of the negative anomalies larger than
©.032 dyne.
Since the publication of the results of fhe recent investigation
of gravity and isostasy, data have become available in regard to a
number of gravity stations established in the Pacific Coast states,
during the summer of 1916. Of 13 stations established in southern
California, each one has a negative anomaly, and the mean with
regard to sign is —0.037. The largest one is —o.081.. This is
only slightly smaller than the anomaly of —o.093 at the Seattle
station. Each of these stations in southern California is located
on Cenozoic material.
There were g stations established during 1916 close to Seattle
with 8 of them on the shores of Puget Sound. Eight of these
stations were on Cenozoic formation, and 7 of these had negative
anomalies. The mean anomaly with regard to sign for the stations
in this vicinity which were established in 1916 is —0.033 dyne.
The writer does not wish to be understood as asserting that the
Cenozoic or the pre-Cambrian material is the cause of the anomaly
at the stations located on those formations. He does believe, how-
ever, that the abnormal density of the material of those two forma-
tions is the cause, or rather the principal cause, of the tendency of
the gravity stations located on them to have anomalies of one sign.
As was stated earlier in this paper, it is not possible from the data
now at hand to tell whether or not the area covered by a Cenozoic
or pre-Cambrian formation, where the sign of the anomaly agrees
with the density of the surface material, is in isostatic adjustment.
This is due to the fact that the compensation, if present, is so far
from the station that its attractive effect is very small in comparison
OUR PRESENT KNOWLEDGE OF ISOSTASY 441
with the effect of the deficiency in density in the materials close to
the surface and under the station.
It seems probable that we may be able to predict with some
accuracy a gravity anomaly on a Cenozoic or pre-Cambrian forma-
tion when we know the latitude and elevation of the point of obser-
vation and make a correction for the topography of the world and
its compensation and apply a correction for the negative or positive
attraction of the deficiency or excess of matter in the Cenozoic or
pre-Cambrian formation. This, of course, is with the provision that
the approximate depth and the horizontal extent of the material of
these formations in the immediate vicinity of the station are known.
It is also possible that where a station is located on a Cenozoic
formation and has a positive anomaly the Cenozoic material is
of slight thickness and is underlaid by pre-Cambrian or other extra
dense material.
The stations in the United States on Paleozoic formations show
a tendency to have negative anomalies. The mean anomaly with
regard to sign is —o.009 dyne. The Mesozoic stations have a
tendency to be positive with a mean anomaly with regard to sign ,
of to.o11 dyne. ‘There were so few stations on intrusive and effu-
sive formations that it is believed that no definite results were ob-
tained from a study of them. There were enough stations in the
pre-Cambrian, Cenozoic, Mesozoic, and Paleozoic formations to
enable one to state rather definitely that stations on any one of
them have a decided tendency to have anomalies of a certain sign.
There cannot, however, be any relation between the sign of the
gravity anomaly and the density of the Paleozoic or the Mesozoic
material, for, in general, the density of the material of those two
formations is about normal.
Where the surface density is subnormal and the gravity anomaly
is positive, it may be possible that there is denser material somewhat
lower down in the lithosphere if the size of the anomaly is large.
By large is meant somewhat above the average size of the anomaly
without regard to sign. There are a number of places in the
United States in which borings have disclosed the presence of
crystalline rocks at varying depths below the surface where the
442 WILLIAM BOWIE
surface material was light in density. Dr. David White, chief
geologist of the United States Geological Survey, suggested to the
writer that it may be possible to predict with considerable accuracy
whether or not crystalline rocks are close to the surface under a
station located on surface material of light density by considering
the size and sign of the anomaly.
There has been considerable confusion in regard to the opinion
of Professor Hayford and the writer as to the density of material
between sea-level and the depth of compensation. Some assert
that we hold that the density is 2.67 for all of this material except
in so far as it is modified by the compensation. As a matter of
fact, neither of us has made any assumption as to the absolute
density of the material between sea-level and the depth of compen-
sation. In the investigations of isostasy it is not necessary to know
the absolute density in making computations of the effect of the
compensation of the topography. It is the deviations ftom normal
densities that are given sole consideration.
A great deal of the gravity anomaly is eliminated by the applica-
tion of the effect of topography and compensation, but it is certain
that the remainder of the anomaly cannot be eliminated by applying
the actual density of the topography in the computations. The
density of 2.67 was used for all the land topography, while there
are local variations in the density of material amounting to 10 per
cent or more. It is, however, impossible that the true densities,
if applied, could have reduced materially the average gravity
anomaly. There is not enough topography to account for the
gravity anomaly. It is of course probable that in many cases the
anomaly would be slightly changed if a true density were used.
What applies to the density of the topography will of course apply
also to the density of the compensation. There is not enough
compensation, however distributed, to account for most of the
anomalies. As the effect of the compensation is the opposite of
that of the topography, the resultant effect is smaller than the
effect of the topography alone. - We must conclude that no method
of distributing compensation applied generally to the country or
to the world will eliminate the gravity anomalies which we
now have.
OUR PRESENT KNOWLEDGE OF ISOSTASY 443
We must go below sea-level and below the beds of the oceans
to find the cause of the anomalies. This necessarily takes the
geodesist into the realm of geology, and it is there that he needs
the assistance of geologists who are familiar with the geological
history of the outer portions of the earth’s lithosphere and of the
existence of materials that deviate from normal density.
As to the process by which isostatic adjustment occurs, we must
consider this largely a matter of speculation. There is no geodetic
evidence on the subject. No one can say that he knows. Of many
theories or opinions one is inclined to accept that which appears
to be most reasonable to him.
We may summarize the contents of this paper as follows: We
have sufficient geodetic data to prove that, for large areas, such as
that of the United States, considered as a whole, the condition of
isostasy is nearly perfect. ‘The data also prove that the local devia-
tions from perfect isostasy are not more than about 25 per cent
oan an average. If, however, we consider that the abnormally
heavy or light material which is found under a number of gravity
stations is compensated for by deficiencies or excesses of density
lower down in the lithosphere, we may assume that the deviation
locally from perfect isostasy is of the order of 10 or 15 per cent
rather than of 25 per cent. The writer believes that this assump-
tion is justified.
There is no geodetic evidence to show whether or not regional
distribution out to a distance of 58.8 km. from a station is more
probable than the local distribution immediately under each topo-
graphic feature. There is geodetic evidence which makes the
local distribution of compensation or the distribution regionally
within 58.8 km. more probable than the regional distribution out
to a distance of 166.7 km. from the station. It may be possible,
though the writer believes it improbable, that there is a distance
between 58.8 km. and 166 km., which would give a more probable
regional distribution than the distances tested.
The geodetic evidence favors about 96km. as the depth of
compensation if the compensation is assumed to be distributed
uniformly between the earth’s surface or sea-level and the depth
of compensation. There is no geodetic evidence to show that any
AAA WILLIAM BOWIE
one method of distribution of the compensation is more probable '
than other methods. It is reasonably certain that any method of
distribution of compensation must have the effective depth of the
compensation at a distance of from 30 to 50 km. below sea-level.
By effective depth is meant such a distance below sea-level that the
effect of all the compensation condensed into a thin layer at that
depth would be the same as the attraction of the compensation
distributed from the surface to the depth of compensation.
The absolute density of the material between the depth of
compensation and the surface of the earth is not considered by the
geodesist in making the corrections for the effect of compensation.
It is only deviations from normal density that he considers, and it
is not necessary even to know what the normal density is.
It must be concluded that the cause of the gravity anomalies
is located below sea-level, and the evidence points to the probability
that at least a large part of the anomaly is due to extra heavy
and extra light material in the outer portions of the lithosphere
which is below sea-level.
There have been found decided relations between the sign of the
gravity anomalies and the geological formation. This is evidently
due to the abnormally heavy and abnormally light materials in the
pre-Cambrian and Cenozoic formations, respectively. There were
found relations between the Paleozoic and Mesozoic formations and
the sign of the gravity anomalies, but it is not evident what has
caused this relationship.
There is practically no relation between the character of the
topography and the sign and size of the gravity anomaly by the
isostatic method of reduction. This is a very strong argument in
favor of isostasy because all other methods of making gravity reduc-
tions which do not consider isostatic compensation show most
decided relations between the size and sign of the anomaly and the
character of the topography.
The use of gravity stations in other countries with those in the
United States gave a gravity formula whose constants were prac-
tically the same as the constants of the gravity formula derived in
1912 from data at 124 stations in the United States alone. In the
OUR PRESENT KNOWLEDGE OF ISOSTASY 445
1917 formula there were 358 stations used, 216 of which were in the
United States.
There was not available for the foreign stations such detailed
information as was available for the stations in the United States,
and it was therefore not possible to utilize the foreign stations in
making certain tests, but the geodetic evidence available for the
foreign stations makes it practically certain that isostasy is in as
nearly a perfect state in those countries as it is in the United
States.
There is no geodetic evidence disclosing the process by which
the isostatic adjustment takes place. This is a matter for specula-
tion rather than proof.
The subject of isostasy is a very important one and a very broad
one, and the work that has already been done is very small in com-
parison with what must be done in order to discover the laws of
the distribution of compensation, the extent to which it is perfect,
and the cause of the unexplained deflections of the vertical and the
anomalies of gravity. The field is broad, and it is necessary that
other scientists than geodesists should enter it. It is especially
desirable that geologists and geophysicists assist in the investigation.
THE SATSOP FORMATION OF OREGON AND
WASHINGTON
J HARLEN BRETZ
University of Chicago
The name ‘“‘Satsop”’ was given by the writer’ in 1915 to a
deposit of stream gravels in the Chehalis valley of western Wash-
ington. The deposit was known then to extend throughout most
of the length of this valley and to occur only in dissected terraces of
stained and decayed gravel standing high above the valley floor.
So far as then known, the Satsop formation rested unconformably
on Eocene and Miocene marine sediments. Because of this rela-
tionship and because of its limitations as a valley filling, it was
thought to be of Quaternary age.
Two field seasons have since been spent in the study of this
formation in Washington and Oregon. It has been found in places
along almost the entire Pacific coast line of Washington and along
the Columbia River valley from the Pacific to the great lava plain
east of the Cascade Range. It has been identified from the litera-
ture along the coast of Oregon. Its relations to the Coast Range
and the Cascade Range are very different, and constitute the chief
reason for the appearance of this paper.
The Satsop formation in the Chehalis valley.—The river gravels
which constitute the Satsop formation of this valley exist along
the lower 60 miles of its total length of 85 miles. The formation
extends back up several tributary valleys, the type sections occur-
ring in one of these, the Satsop valley. The maximum known
thickness is 300 feet. The formation is composed of local materials
and is stream-bedded with dip down the present drainage lines.
Dissection has reduced the formation to a series of terraces, and
decay has produced a residual loam on the surface of the highest
1] H. Bretz, ‘‘Pleistocene of Western Washington,” Bull. Geol. Soc. Am., XXVI
(to15), 131.
446
THE SATSOP FORMATION 447
terraces and given a dull red or orange color to the upper 50 feet of
gravel.
The Satsop formation along the Pacific Coast of Washington.—
The terrace gravels which constitute the Satsop formation of the
Chehalis valley are traceable almost continuously in the cliffs along
the lower part of this valley and in the bluffs of Grays Harbor to the
sea-cliffs of the narrow coastal plain.
North of Grays Harbor the formation differs from that in the
Chehalis valley only in containing much clay and sand, with frag-
ments of driftwood. In places there are strata of peat or lignite
several feet thick. The gravel in some exposures is a beach shingle
and lies on wave-worn and mollusk-drilled Tertiary sandstone.
The formation is horizontal for the most part, and such warping as
does exist is very slight. The formation rests unconformably on
Tertiary and older rocks. A few marine shells record the presence
of the sea, and the interbedded peat tells of tidal marsh condi-
tions in places during accumulation of the deposit. The thickness
of the formation as shown in the cliff sections does not exceed
200 feet.
The shore line of Willapa Bay, south of Grays Harbor, is
largely cliffed, and all of the cliffs are cut in the Satsop formation.
Clay and sand predominate. Peaty strata record the presence
of fresh- or brackish-water swamps during aggradation. Shells of
marine mollusks, cross-bedding due to tidal currents, and beach
shingle in the gravelly strata tell of deposition in marine water.
One stratum of highly fossiliferous clay is traceable for several
miles along the bluffs. Most of the shells in it are of oysters, many
of the valves yet attached in pairs. The shell-bearing stratum rests
on blue clay, which is full of molluskan borings but contains no
shells. Above the shell bed is a peaty clay containing much driit-
wood. Stumps 7m situ and upright stems in this layer record suc-
cession of the oyster bed by a coastal marsh. In the gravelly
strata are pebbles of granite, gneiss, schist, and quartzite, all but
the quartzite considerably decayed. None of these materials occur
in the drainage area of either the Chehalis or the Willapa River,
while all of them are common in the Satsop formation of the
Columbia valley. This gravel undoubtedly was brought over into
448 J HARLEN BRETZ
the Willapa Bay region by distributaries of the Columbia during
the Satsop aggradation.
The Satsop formation in the Willapa Bay region is a little more
than 75 feet in maximum exposed thickness, with the base below
tide. The strata are horizontal or depart from that attitude only
in gentle undulations.
The formation may be traced back up the Willapa valley into
terraces of a decayed and red-stained river gravel which rests on
eroded Eocene basalt and Miocene sandstone. The relation
between the coastal and valley phases in the Willapa valley is the
same as that in the Chehalis valley.
The Satsop formation along the coast of Oregon.—J. S. Diller*
has described Quaternary sediments along the Oregon coast which
belong clearly to the same formation as those along the coast of
Washington. Diller did not name this formation and it has sub-
sequently received but passing mention in the literature. Hence
the name ‘‘Satsop’’-is here extended to cover that Quaternary
formation of the Pacific Coast whose minimum limits reach from the
Strait of Juan de Fuca north of Washington to the Coquille valley,
within 80 miles of the Oregon-California line.
Exposures of the Satsop formation examined by Diller are as
follows:
Ilwaco, Washington: 14 feet of gravel, sand, and clay, the
top lying 30 feet above tide. Contains fresh shells of living
species of mollusks. Unconformable on tilted shales of Oligo-
cene age.
- Tillamook Bay, Oregon: 20 feet of sandstone, capping a
bluff of basalt 300 feet high.
Yaquina Bay, Oregon: Nye Beach: 40 feet of horizontally
bedded gray sand overlying 20 feet of tilted Miocene shales.
Sand contains logs, branches, and roots, some roots apparently
in situ. Another section: to feet of yellow sand overlain by 5 feet
of indurated gravel, this overlain by 6 feet of sand. Another
section: 30 feet of Quaternary materials containing cones identified
by F. H. Knowlton as of tideland spruce.
J. S. Diller, ““A Geological Reconnaissance in Northwestern Oregon,” U.S. Geol.
Surv., 17th Ann. Rept., Part I, 1896.
449
PoTRac| es
PAH | ca ENING
rs s ipared ee ate Be
cai pe RS ee
Py gata est
ae
uonruoy dos}vs Sa
THE SATSOP FORMATION
Sand and gravel unconformable on Miocene
Clay contains a multitude of marine shells, identified by
Newport Point:
shales.
Wood and cones also
Dall as belonging to six living genera.
present.
Coquille valley: 30 feet of sand, summit 50 feet A.T.
450 J HARLEN BRETZ
The Satsop formation in the Columbia valley west of the Cascade
Range.—The Satsop formation in the lower Columbia valley does
not differ in any essential from that in the Chehalis and Willapa
valleys. It contains a surprisingly large amount of quartzite
gravel. In some strata more than 50 per cent of the pebbles are of
quartzite, all of which undoubtedly have come from east of the
Cascade Range. Basalt, a common country rock, is also a leading.
constituent of the gravel. The basalt pebbles are decayed, except
in the deeper portions of the deposit.
There are three large areas in the drainage of the lower Columbia
where the Satsop formation covers many square miles, instead of
being limited to narrow terraces. One of these areas is in the valley
of the Cowlitz River, a tributary of the Columbia from the north;
a second is in the valley of the Willamette River, a tributary from
the south; and a third lies in a broad portion of the Columbia
valley between the two areas just mentioned.
The Satsop formation in the Cowlitz valley is at least 150 feet
thick. It here constitutes a broad, terraced plain and rises north-
ward to a summit level tract about 500 feet A.T. This tract is a
portion of the divide between the Cowlitz and Chehalis rivers. It
bears a residual soil and with little doubt is part of the original
upper surface of the formation. The pebbles in the upper 30-50
feet are softened by decay, those immediately below the soil being
spaded through in excavating. At depths greater than 50 feet the
pebbles are hard, but the reddish to yellowish stain penetrates as
far as excavations have gone. No quartzite pebbles have been
found in this part of the Satsop formation. The dissection of the
tract is adjusted to a base-level recorded by a broad terrace 100
to 150 feet lower than the summit plain and about 250 feet above
the present flood plain of the Cowlitz River. This terrace has
been found in’ most of the major valleys of the region studied.
From its notable development in the Cowlitz valley it is here
named the Cowlitz Terrace.
Only the lower 25 miles of the Willamette valley of Oregon have
been examined in the study of the Satsop formation. Most of this
portion is covered by the Satsop. Numerous hills of basalt rise
THE SATSOP FORMATION 451
through and several hundred feet above the surface of the Satsop
fill. The formation is at least 600 feet thick along the Sandy River,
with the base below river-level. The material is stream-bedded
gravel and sand, indurated in some places to a conglomerate and
sandstone. Quartzite is a common constituent for ro miles south
of the Columbia, but has not been found more than 15 miles from
the master-stream. Quartzite and basalt are the most important
constituents.
The Satsop formation of the lower Willamette valley is maturely
dissected, the dissection adjusted to a base-level 200 feet or more
above present flood plains. This level is recorded in the major
valleys by a prominent terrace developed mostly in the Satsop
formation but in places cut in the underlying basalt. This is the
Cowlitz Terrace already described.
The uplands of this Satsop plain bear a red clay soil 10-15 feet
deep. This grades down into a much-decomposed gravel. At a
depth of 30 feet the pebbles are decayed only on the exterior.
Below 50 feet most of the material is hard and ringing when struck
with the hammer. Near the Columbia the clayey residual soil on
the top of the Satsop formation contains scattered quartzite
pebbles, hard, bright, polished, and apparently unaffected by the
weathering which has reduced the associated basaltic pebbles to a
structureless clay.
The surface of the Satsop formation in the Willamette valley
lies at about 500 feet A.T. in mid-valley and rises eastward toward
the Cascade Range to 1,200-1,500 feet, at these altitudes passing
under the more recent lava-flows of this range. No upward slope
of the Satsop surface toward the Coast Range on the western side
of the Willamette valley has been found. On this side the forma-
tion terminates against hills of older basalt.
The broadened portion of the Columbia valley between the
Cowlitz and the Willamette is really a continuation of the Willa-
mette valley northward into Washington. The surface of the
Satsop formation constitutes at least 200 square miles of the flat
floor of this part of the valley. It is disposed in two levels, approxi-
mately 300 and 500 feet A.T., the lower of which is the Cowlitz
452 J HARLEN BRETZ
Terrace and the upper probably the original surface of the formation.
This filling is very similar in all features noted in the preceding
description to that in the Willamette valley.
Relation of the Satsop formation to the Coast Range of Oregon and
Washington.—The Chehalis and Columbia rivers cross the Coast
Range in capacious valleys of low gradient. In both valleys
the Satsop formation has been found at closely spaced intervals from
the coastal plain portion to the broad fillings east of the range.
Though not continuous across the range, the character of the
material of the formation, the altitudes at which it occurs, the
stratigraphic relations to underlying rock and to younger gravels,
and the topographic relations are such that there seems no reason-
. able doubt that the deposits noted in this paper are portions of the
same formation.
The Cowlitz, Chehalis, Columbia, and Willamette valleys are
younger than the Coast Range, and the Satsop formation is
younger than the valleys in which it lies. Thus the Satsop forma-
tion was deposited after the Coast Range had been uplifted, and
after its dissection was well advanced toward present maturity.
The Satsop formation along the Columbia River in the Cascade
Range.—The Columbia has cut its valley across the Cascade Range
down almost to sea-level. This valley is a gorge about 60 miles
in length, only the western 35 miles of which have been mapped
topographically. Most of the walls of the gorge are precipitous
and maximum sections of 4,000 feet are available.
Numerous bluffs along the lower 12 miles of the Oregon side of
the Columbia Gorge reveal a flow of gray basalt, 25-100 feet thick,
in the Satsop formation. The Satsop rises northeastward in the
walls of the gorge about go feet to the mile, bringing its base goo
feet A.T. in the salient known as Angels Rest (Fort Rock) and
exposing the Columbia River lava below the Satsop. Many sections
along this distance show an unconformable contact between the Sat-
sop and the underlying basalt, and some of them show the upper
10-20 feet of this basalt to be very much decayed, far exceeding the
decay of the basaltic pebbles in the lower part of the Satsop. In
the section at Angels Rest the Satsop (including the intra-
formational lava) is 500 feet thick. A second lava-flow appears in
THE SATSOP FORMATION
this section, capping the Satsop gravel and con-
stituting the surface formation back from the
edge of the Columbia Gorge. It may be traced
along the cliffs to a volcanic cone known as
Devils Rest, overlooking the gorge but a little
back from the bluffs. Devils Rest is one of
a dozen or more such cones near the gorge
which have supplied the lavas overlying the
Satsop formation.
The accompanying section (Fig. 2) tells the
rest of the story better than would a detailed
description. The intra-Satsop flow does not
appear elsewhere in the range, but volcanic
fragmental material is prominent in most sec-
tions of the formation. All phases of this
material are present, from ash and lapilli to
volcanic bombs, rudely stratified by their fall;
from slightly water-rolled and poorly sorted
débris to well-worn pebbles and cobbles of
the gray lava, associated with equally worn
pebbles of Columbia River lava and beautifully
smoothed and polished pebbles of quartzite.
The Satsop deposit invariably rests on
eroded Columbia River lava (a dense black
basalt) and is capped by flows of gray basalt.
In places the Satsop is absent and the two
lava formations are in contact. The highest
altitude at which the Satsop formation has
been found in the Cascade Range is 3,700 feet
A.T. in Benson Plateau midway across the
range. Its thickness here is nearly 700 feet.
On the eastern slope of the main range the
river phase of the Satsop (i.e., well-stratified
gravel composed predominantly of basalt and
quartzite) appears at 2,500 feet and descends
eastward to too feet A.T. in the synclinal
Hood River valley. The formation in this is
JOATY
= pooy
neaqeg
uosueg
JOA
ApuesG
JOA
OIJOUIETIIM
453
The Dalles
Portland
Looking north. Satsop formation shown in black
Oregon side.
Fic. 2.—The Columbia River section of the Cascade Range.
454 J HARLEN BRETZ
strikingly like that along the Sandy River, just west of the range.
This is true of its composition, its structure, its degree of weathering
and of cementation. It differs in having no lava-flow within the
formation.
Between Hood River and The Dalles, the Columbia River has
cut across two anticlinal ridges, each rising more than 2,000 feet
above the stream. Each anticline is composed almost wholly of
Columbia River lava. Each carries patches of a sedimentary
formation composed chiefly of volcanic detritus, but containing
much rounded gravel in which basalt, granite, and quartzite are
present.
In the vicinity of The Dalles is a stratified deposit of volcanic
agglomerate, tuff and ash, with strata of river sand and gravel,
1,000 feet thick, and capped by a flow of gray basalt. The western
margin of the deposit is uptilted on the flank of the eastern anti-
cline. Though no pebbles of quartzite or granite were found
during the brief examination possible, it seems probable from
stratigraphic evidence that the deposit at The Dalles is a local phase
of the Satsop formation.
The Satsop formation between the Columbia and V akima valleys.—
The Simcoe Range is a prominent eastward spur of the Cascades,
extending some 50 miles east of The Dalles, and separating the
lower Yakima valley from the Columbia valley to the south. The
range is structurally a broad anticlinal fold. Typical Satsop
quartzite gravels, resting on dense black basalt and covered by
gray basalt, lie in many places on the southern slope. The highest
altitudes at which these gravels are known to occur in the Simcoe
Range is 4,000 feet. The overlying lava does not extend far down
the northern slope of the range, and the Satsop formation on this
slope, unprotected by a lava cap, consists only of scattered quartz-
ite cobbles and pebbles. All other materials in the original deposit
have been destroyed by weathering. Quartzite pebbles were
found as far north of the range as the southern part of the Ellens-
burg quadrangle. There are areas within sight of the Yakima
River where these pebbles cover 50 per cent or more of the surface.
The Yakima basalt in the Ellensburg quadrangle (probably the
equivalent of the Columbia River lava) is overlain by the Ellens-
THE SATSOP FORMATION 458
burg formation, largely a sandstone of volcanic débris. Both the
Yakima and Ellensburg formations have been folded into a number
of east-west anticlinal ridges. Ahtanum Ridge is one of these
folds in the southern part of the Ellensburg quadrangle. On this
ridge the quartzite pebbles lie on the eroded edges of both the
Yakima basalt and the Ellensburg sandstone. ‘Toppenish Ridge,
between the Ahtanum and Simcoe anticlines, is similarly oriented,
of similar origin, and bears abundant quartzite gravel on the edges
of both formations.
Relation of the Satsop formation to the Cascade Range.—It is
obvious from data already presented that the outpouring of gray
basalt immediately succeeded, and in part was contemporaneous
with, the deposition of the Satsop formation in the Cascade Range.
From the position of the Satsop formation in these mountains, it is
also clear that it and the overlying lava-flows were deposited before
the Cascade Range was formed.
Relation of the Satsop formation to the Methow peneplain.—
Russell first advanced the hypothesis that the accordant summit
levels of the Cascade Mountains in central and northern Washington
record a warped and dissected peneplain. Willis and Smith' have
named this the Methow peneplain. They have identified it on the
eastern slopes of the Cascades from Lake Chelan on the north to
the Yakima valley on the south. In the Ellensburg quadrangle the
Methow pleneplain is thought to truncate the Ellensburg sandstone
and the underlying Yakima basalt. As interpreted by Smith, these
formations were gently folded before the peneplanation. Develop-
ment of the peneplain brought the surface of these folds down to
base-level. Renewed folding along the same axial lines is thought
to have followed the truncation so that the Methow peneplain is
now a warped surface lying on the tops and flanks of the anti-
clinal ridges.
The significant item here contributed is that the mantle of
Satsop quartzite pebbles lies unconformably on the tops and flanks
of at least some of these anticlinal ridges. If they constituted a
stratified deposit across the eroded edges of the underlying forma-
tions, the case for planation between two epochs of folding would
t Bailey Willis and George Otis Smith, U.S. Geol. Surv., Prof. Paper 19; 1903.
456 J HARLEN BRETZ
be complete. However, the position of the mantle of loose pebbles
is so strikingly similar to that of the stratified Satsop on the southern
flank of the Simcoe Range and throughout the Cascade Range that
ittle hesitation is felt in correlating the eroded surface named the
‘“‘Methow peneplain”’ with the eroded surface beneath the Satsop
formation in the Cascade Range.
But this eroded surface as exposed in the Columbia Gorge is
irregular and numerous elevations in it rise several hundred feet
above the base of the Satsop. This is well shown in the Willamette
valley and the Hood River valley where these hills of basalt rise
through the Satsop formation and the younger lavas. The surface
on which the Satsop rests in the sections of the Columbia Gorge
may be post-maturely eroded, but it is not a peneplain.
Further, it has been shown in this paper that the Coast Range
rises above the Satsop formation in the Chehalis, Willapa, Willa-
mette, and lower Columbia valleys and therefore was not a pene-
plain at the time of Satsop deposition.
The age of the Satsop formation.—Diller’ reports that W. H.
Dall found all of the species in a collection of shells from the
Quaternary deposits on Yaquina Bay, Oregon, to be living forms.
He also states that F. A. Lucas identified a large tooth from the
same beds as that of a Pleistocene mastodon and that F. H. Knowl-
ton identified cones from this formation as those of “‘ Picea stichensis
Carr.,” the tideland spruce growing along this coast today.
In another paper’ Diller has described deposits between Cape
Blanco and Elk River, Oregon, about 50 miles north of the Cali-
fornia line, which he names ‘‘Elk River beds.’’ Dall states that
the fossils collected from these beds are probably Pleistocene in age.
Martin’ has more recently examined the Cape Blanco region and
reports two faunal horizons in the Elk River beds, the upper of
which is very closely related to the Upper San Pedro series and “‘is
rJ.S. Diller, ‘‘A Geological Reconnaissance in Northwestern Oregon,” U.S. Geol.
Surv., 17th Ann. Rept., Part I, 1806.
2J. S. Diller, “‘The Topographic Development of the Klamath Mountains,”
U.S. Geol. Surv., Bull. 196, 1902, p. 30.
3 Bruce Martin, ‘‘The Pliocene of Middle and Northern California,” Univ. of Cal.
Publications, Bull. Dept. Geol., IX, No. 15 (1916).
THE SATSOP FORMATION — 457
at least Pleistocene in age”’ and the lower of which probably is very
late Pliocene. The Elk River beds overlie the Empire beds with
angular unconformity. Ralph Arnold and B. L. Clark’ consider
that the fauna of the Empire beds is “‘of very nearly the same age
as that of the Purisima formation in the Santa Cruz Mountains
of California, which is Pliocene, and not the oldest Pliocene”’
(Clark). The Elk River beds are apparently the same as the
deposits of the Oregon coast farther north, which Diller called
Quaternary.
Ralph Arnold,” in a description of the geology of the coast of
the Olympic Peninsula of Washington, maps and names what is
here called the Satsop formation as ‘“‘ Pleistocene gravel, sand, and
clay.” He notes the presence of tilted Pliocene beds (his Quinault
formation), bearing a fauna similar to that of the Purisima forma-
tion of California. On these beds the Satsop formation rests with
angular unconformity. B. L. Clark’ believes that present knowl-
edge of the Pliocene faunas of the Pacific Coast upholds Arnold’s
determination and that the scarcity of extinct species suggests
strongly that the fauna is rather late Pliocene in age, though not
the latest Pliocene.
Harold Hannibal, in a paper by Ralph Arnold and himself,
notes the Quaternary age of the oyster-shell bed in the Satsop
formation of the Willapa Bay region.
Diller also collected fossil shells from shales 700 A.T. on the
slopes of the Columbia valley 35 miles from the coast. Dall refers
the shells to the Pliocene. The Satsop here is a terrace gravel down
in the valley and younger than the Pliocene beds.
A clay stratum with abundant fossil leaves has been found in the
Satsop formation on the western slope of the Cascade Range.
Knowlton has examined collections from this bed and is of the
opinion that the flora is Quaternary in age. He finds leaves of
t Personal communication.
2 Ralph Arnold, ‘A Geological Reconnaissance of the Coast of the Olympic Penin-
sula, Washington,” Bull. Geol. Soc. Am., XVII (1906), 451.
3 Personal communication.
4 Arnold and Hannibal, ‘‘The Marine Tertiary Stratigraphy of the North Pacific
Coast of America,” Proc. Am. Philos. Soc., 1913.
458 J HARLEN BRETZ
Quercus venustula and Sequoia sempervirens in the collection. Both
are living species.
Age of the Cascade Mountains.—Ii the foregoing determinations
are correct, the Cascade Range, at least in this portion, is of
Quaternary age.
Acknowledgments.—During the progress of the field work on
which this paper is based, nine weeks were spent with the Wash-
ington Geological Survey and four weeks with the Oregon Bureau
of Mines and Geology. I wish to acknowledge indebtedness to
both organizations, and in particular to Mr. Ira A. Williams, with
whom I was associated during the work for the Oregon Bureau.
THE CORROSIVE ACTION OF CERTAIN BRINES
IN MANITOBA’
R. C. WALLACE
University of Manitoba, Winnipeg, Canada
The brines geologically considered.—The Manitoban escarpment
consists of a range of hills which fringes on the western side the
lake system of which Cedar Lake, Red Deer Lake, Lake Winni-
pegosis and Lake Manitoba are the most important members, and
extends southward beyond the international boundary line into
North Dakota. It reaches, in the Porcupine Mountain, a maximum
elevation of 2,500 feet above sea-level. At the foot of the escarp-
ment the plain slopes gently eastward from an elevation of goo
feet to one of 700 feet. The escarpment is the eastern edge of the
Cretaceous shales which extend throughout the western prairies.
The shales were uplifted in early Tertiary times, and were eroded
from the Red River valley back to the escarpment before the end
of the Tertiary period. On this surface of erosion limestones of
Paleozoic age are exposed, Ordovician, Silurian, and Devonian
strata appearing successively from the edge of the pre-Cambrian
shield to the escarpment; while the Dakota sandstone—the lowest
member of the Cretaceous series—rests directly on the surface of
the Devonian limestones. The basin of the lake system has been
carved from Devonian limestones and dolomites.
At the foot of the escarpment, on the west side of the lakes, a
series of salt springs emerges from middle and upper Devonian
strata (the Winnipegosan dolomite and Manitoban limestone
respectively).2 These springs follow the base of the escarpment
for a distance of 250 miles, but are found in greatest numbers
on the west shore of Dawson Bay, at the north end of Lake
t Published with the permission of the Directing Geologist, Geological Survey of
Canada.
2R, C. Wallace, ‘Gypsum and Brines in Manitoba,’ Memoir Geol. Surv. Canada
(in press).
459
460 RCM WAL PAGE
Winnipegosis. Around the springs are flats, absolutely devoid of
vegetation, from a half-acre, as the case may be, to several hundred
acres in extent. These the traveler may quite unexpectedly find
in the midst of a dense forest; but the majority of the springs are
to be found in close proximity to river or lake.
The brines are not confined to a single geological hoezor
They appear in both the Winnipegosan dolomite and Manitoban
Fic. 1—Bowlder-strewn salt flat, Geikie’s Creek, Sagemace Bay, Lake Winnepegosis
limestone, the combined thickness of which is approximately 250
feet. Owing to the comparatively level surface and the appre-
ciable dip of the strata toward the west, the difference in elevation
between the various springs is very small—not more than 50 feet.
The springs may consequently be referred with greater exactness
to a contour horizon than to any geological horizon.
The Dakota sandstone, which directly overlies the compara-
tively porous Devonian strata, is a well-known water-bearing
CORROSIVE ACTION OF BRINES IN MANITOBA 461
horizon in the middle northwestern states and in the western
provinces. It is capped by impervious shales, and the water,
which circulates under considerable hydrostatic pressure, appar-
ently penetrates laterally into Devonian strata, leaches sodium
chloride from certain horizons in which the salt has been precipi-
tated with the limestone, and reaches the surface where the cover-
ing of drift is thin or absent. On an average approximately 430
gallons of brine reach the surface per minute during the dry season;
and the salt, if evaporated, would cover the main salt area (200
miles by 30 miles) with a coating 2 feet thick in 10,000 years.
TABLE I
Brine* Sea-Water{ Brine* Sea-Watert
Kereta) sans 37, Tenor HCO OW DOM HY se Nan Nya) re
ING eee 34.99 30.50 OE AN AG. 55-95 55-29
Gare ste sci 2.02 I.20 Brea nee vs: 0.04 0.19
IN eee 3 oun 0.55 B73 Pi eeha rac DIT Bs eV SA ea
TE ees ee ee aac msn aa SS ENE Re ee ne STAM eva OPOZa NN NIAMS MCE Net
HN Yate AEs aN ONO TAME Cu RCNa M o ts Percentage
SOP ee 4.88 7.69 salinity... 7.29 3.30 to 3.74
COPE ene Nil 0.21
* From Salt Creek, Salt Point Peninsula, Lake Winnipegosis. Professor M. A. Parker, analyst.
t Mean of 77 analyses by W. Dittmar.
Composition of the brines—Numerous analyses have been made
of the brines. The composition is remarkably uniform, differences
occurring only when the brines pass through a considerable depth
of glacial drift before reaching the surface. In this case the per-
centage of Ca and SO, ions is considerably greater than normal.
Table I gives the analyses of a typical brine, Dittmar’s average
of 77 analyses of sea-water being given for comparison.
The analyses are given in percentages of total solids. The per-
centage of salinity in the analysis quoted is somewhat greater than
normal, but the percentage values for the constituents vary only
slightly from the figures quoted. While on the whole the brine is
very similar to sea-water, it is a distinctly purer solution of sodium
chloride. The relative percentages of Ca and Mg ions differ in
a sense, which may be accounted for by the abstraction of Ca ions
by marine organisms. ‘The apparent differences in the carbonate
462 RSC) WALLACE
values are due to the fact that in the statement of the analysis of
sea-water the total carbonate is reckoned as normal carbonate.
The concentration of the brine is, however, notably higher than
that of sea-water.
Action on the bowlders.—The glacial drift was to a large extent
derived from the pre-Cambrian areas in the north. The bowlders
which cover the bare flats where the salt springs are found are
mainly gneissose or granitic, though occasionally dark-green epi-
diorites or Paleozoic limestones are seen. Chemical solution has
taken place on an extensive scale, many bowlders having been
reduced on every side by at least a foot. This is very clearly seen
in the salt creeks in which the water is carried from the springs to
the lake, the bowlders standing on a much-eroded base, like the
rocks of a great sand desert. On the flats the bowlders are pitted
into very fantastic forms, the ferromagnesian minerals having
suffered to the greatest extent. Not even quartz nor garnet has
escaped the action of the solvent. Gneissose structures are accen-
tuated by differential weathering, garnetiferous bands standing out
in special relief. As the corrosive power of the brine is apparently
much more intense than that of sea-water, it is of interest to inquire
into the processes involved in the disintegration of the rock.
Chemical processes.—Regarded as a chemical agent, the brine
may be considered to be a weak solution of sodium chloride. A
considerable amount of experimental work has been done on the
action of solutions of sodium chloride on minerals, but the evidence
is somewhat conflicting.t Joly has, however, proved that sodium
chloride, in the presence of the atmosphere, is a more active dis-
integrating agent than pure water. Daubree’s experiments were
conducted under somewhat different conditions. In the case of
the brines physical conditions have been favorable. Normally
the salt crystallized in thin crusts at the base of the bowlders. The
salt is somewhat deliquescent, and gradually extends upward over
the side of the bowlder, till a thin coating of brine, somewhat diluted
during the process, covers the whole bowlder. The conditions are
thus most favorable for chemical activity in presence of the atmos-
tDaubree, Synthetische Studien zur Experimentalgeologie, 1880, 205; Thoulet,
Compt. Rendu, CX, (1890), 652; Joly, Proc. Roy. Irish Acad., XXIV (1902).
CORROSIVE ACTION OF BRINES IN MANITOBA 463
phere; and the writer believes that it is primarily as an agent for
distributing the liquid in a thin film over the bowlder, and only
secondarily as a direct chemical agent, that the dissolved material
in the brine acts in the process of disintegrating the bowlder. It
has been proved conclusively that water is itself an agent of con-
siderable chemical power and that it acts most vigorously as a
Fic. 2.—Corroded bowlder, salt flat, Pelican Bay, Lake Winnipegosis
corrosive agent when in intimate contact with the atmosphere,
as, for instance, at a water surface.
The actual process of disintegration is necessarily different for
different minerals. The ferromagnesians, more particularly the
amphiboles and pyroxenes, have suffered to a greater extent than
the feldspars. The alkaline earths are somewhat readily attacked
and dissolved as carbonates or chlorides, and silica with alumina,
mixed or combined, is left in colloidal form. The percentage of
soluble material in the case of the feldspars is correspondingly
smaller. To some extent, with the metasilicates at least, the
464 R. C. WALLACE
process is one of hydrolysis; and while it may be effected by water
alone, it is doubtless hastened by the carbon dioxide of the atmos-
phere. The gels which are formed during the process of decom-
position are irreversible—that is, they cannot, by the action of
electrolytes, pass over into sols and be in this way removed from
the sphere of action. They exercise, however, a selective absorp-
tion, alkalies being removed from the brine while the acid radicals
remain in solution. With potassium salts in particular this
property of the colloids of the soils is of importance in retaining
the valuable ingredients of fertilizers. This selective absorption
tends, therefore, to hydrolyze the chloride, and to render the solu-
tion more acid. ‘The free acid reacts on the partially decomposed
minerals, causing further disintegration. Quartz is not affected
thereby, but the corrosion of quartz is in all probability due to the
action of alkaline carbonates.
In short, then, the principal fact in the disnteeauon is the
intimate contact of the liquid (in a very thin film) with the atmos-
phere and the rock. The initial stages of the disintegration are
caused by the action of water in contact with air rather than by
that of salts in the water. Colloidal precipitation, which takes
place when decomposition of the silicate begins, leads to selective
absorption and consequent acidification of the solution, giving rise
in turn to further, and probably more intense, disintegration. The
process is continuous, gel being continuously precipitated, and
further selective absorption taking place. The gel, being irre-
versible, is not taken up as an emulsion, and can consequently be
removed only mechanically.
Comparison with the action of sea-water—On comparing the
action of sea-water on bowlders of similar composition to those
attacked by the brines, one finds an undoubtedly real difference
in the degree of corrosion. It is, however, a difference in degree, not
in process. ‘The evidences of chemical erosion caused by the sea-
water are to a large extent removed by mechanical attrition caused
by the impact of the waves on the bowlders, and the consequent
rolling of the bowlders on the beach. Even in large bowlders,
however, where rolling is reduced to a minimum, the evidence of
chemical disintegration is small indeed in comparison with that of
CORROSIVE ACTION OF BRINES IN MANITOBA 465
4
the bowlders from the salt flats. The relatively small difference
in concentration of sodium chloride is not sufficient to account for
the difference in chemical activity in the two cases.
Bowlders between high and low watermark are situated simi-
larly to those on the salt flats in one particular—they are in intimate
contact with the solution and with the atmosphere. They are
Fic. 3.—Corroded bowlder, salt flat, Pelican Bay, Lake Winnipegosis
differently placed, however, in that they are alternately water-
covered and dry to the base. Evaporation does not proceed so
far that sodium chloride is precipitated, and films of liquid are not
fed over the surface of the bowlder from the base. The initial
disintegration of the bowlder is consequently less readily effected,
and the acidification of the solution through absorption is to a
similar extent retarded.
While to different physical conditions may be attributed the
difference in degree of corrosion of beach bowlders and bowlders
466 RC: WAELACE
of similar type in salt flats, the fact of chemical erosion by sea-
water is emphasized by the study of the action of this essentially
similar brine. The chemical action of sea-water on igneous rocks
between high and low watermark must be much more considerable
than is generally believed. The disintegration attributed to
mechanical attrition is undoubtedly, in part, at least, chemical.
Sea-water under great pressure is apparently a solvent for the
volcanic débris which reaches the bottom of the deeper ocean; but
the solvent power is intensified by contact with the atmosphere,
even at ordinary pressures. Conditions are most suitable when
the three phases—solid, liquid, and gas—remain in intimate con-
tact for considerable periods of time. Such is the case where
shallow pools of water are imprisoned in the hollows of the rock
surfaces when the tide recedes; in these cases evidences of corro-
sion are very clear.
It would be futile to attempt to compare in intensity the action
of sea-water on beach bowlders and that of rain-water impregnated
with humus acids from the soil. Data are not available in the
field. It must suffice, at this stage, to rank sea-water and acidified
rain-water side by side as two potent agents from the chemical, as
from the mechanical, standpoint, in the disintegration of rocks.
NOTES ON THE 1916 ERUPTION OF MAUNA LOA
HARRY O. WOOD
Hawaiian Volcano Observatory
Il
The writer spent the two days and night of May 30 and.31, 1916,
in making a hurried reconnaissance of the Kahuku branches of the
recent flow, and of the region near its source, where action was still
going on, though the eruption already had greatly diminished.
There had been no forward movement of the flows at their fronts
later than May 27 or May 28. A preliminary account of this
work was published at once in the Weekly Bulletin of the Hawaiian
Volcano Observatory.t However, difficult foot traveling was
encountered, and somewhat adverse weather conditions, which
prevented thorough and accurate work; and the presence of fog
and fumes led to erroneous estimates of distances and heights;
also, our guide applied place-names incorrectly and his errors
naturally found place in the early report. Further, observations of
much interest, but not pertinent to the present subject, were
included in that account. Hence an abridged and corrected state-
ment relating to the 1916 action finds appropriate place here.
Much that was only glimpsed on this hurried trip was fully con-
firmed by the later study. Whence certain observations thus
confirmed it is convenient to mention here.
ITINERARY
The writer and one companion left the observatory in the early
evening of May 29 and motored to the village of Waiohinu, where
the night was spent. Early the next morning we motored to a
gate of the Kahuku ranch about a mile west of the ranch buildings.
Here we were joined by the guide, and the party set out on horse-
back at 8:00 A.M., going up the south slope of the mountain between
t Vol. IV, 6, pp. 51-57.
407
408 HARRY O. WOOD
the flows of 1868 and 1887. At 10:30 A.M. we reached the southern-
most ‘‘toe’”’ of the front of the Kahuku branch of the 1916 flow.
Here we turned off to the right to pass around this flow and up on
the east and north of it. We followed an upland trail toward
Kapapala until we came into a Kipuka, a long, narrow strip of
forest land extending up the mountain between two barren streams
of a—a, known by the name Kipuka Akala. At the lower end of
this we left our horses. We reached it shortly before noon, and
at noon we set out on foot up the mountain, taking a northwest
course toward the general source of the new flow, north of the con-
spicuous cluster of cinder cones marked Puu o Keokeo on the
government map. (There is dispute as to whether this name is
correctly applied to this group of cones, but there is no doubt that
the members of the group so mapped were those identified.)
As we planned to spend the night near the source we were
necessarily laden with food, water, photographic equipment, etc.,
and blankets or extra clothing—a moderate load for each man.
For about an hour we made our way upward alongside the strip of
forest, but walking in the open chiefly over old a-a. At about
1:00 P.M. we came out onto a barren, complex network of a—a flows
of varying direction and age—much of this apparently of 1907
date—and over this very difficult surface we clambered until a
little after 5:00 P.M. We then had reached an old cinder cone ~
situated between two and three miles from Puu o Keokeo in a
direction a trifle east of north. This cone was about a quarter of
a mile east from the rift-line source of the 1916 flow described below.
We had not reached the upper limit of this source, which had been
our goal, but it became impracticable to go on farther. Here we
stationed ourselves in the lee of this old cinder cone and passed
the night, practically all of which the writer devoted to observation
of the action and conditions along the visible length of the rift
source.
At 4:50 A.M. on May 31 we began the descent. At 9:40 A.M.
the horses were reached, and at 10:15 A.M. we started on our way
to Puu o Keokeo, going around the southern end of the Kahuku
branches of the flow and up on the western side of them. By noon
we reached a flat clearing, where we dropped our camp equipment
NOTES ON THE 1916 ERUPTION OF MAUNA LOA 469
and immediately proceeded upward on horseback to a point very
near to Puu o Keokeo, on a low ridge of ancient pahoehoe which
extends a short distance eastward from the conspicuous cluster of
cinder cones. In clear weather this slight eminence affords an
expansive view to northward, and much wider views can be obtained
from the summits of the cones near by. On this occasion we
obtained only partial views through driving fog and clouds, and in
brief clear spells between small local showers. We reached this
point of outlook at 2:20 p.m. It was impracticable to keep horses
here overnight, and unnecessarily uncomfortable to remain our-
selves without camp equipment in such unsettled weather, espe-
cially as our hasty reconnaissance was completed, so we began our
descent at 2:35 P.M., and returned on horseback to Waiohinu,
which we reached at 8:30 P.M. We returned to the observatory
the next day.
Thus we passed completely around the Kahuku branches of
the new flow, except for their breadth near the source. In this
way, especially on the foot journey, and also from the ridge near
Puu o Keokeo, a good general survey was obtained of conditions
along these and near their source; and in limited localities, particu-
larly those encountered on foot along the eastern side of them,
many matters of detail were observed. Conditions of the traveling,
distance, and weather prevented examination of the upper limits of
the source, of the action and conditions on the sides toward Kona,
and the making of photographic records.
GEOGRAPHICAL NOTES
The closely grouped cluster of old cinder cones, named Puu o
Keokeo on the government map, of which the highest point is
6,870 feet above sea-level, forms a low but conspicuous landmark
on the south-southwest flank of Mauna Loa. Extending east-
southeastward from this for a half-mile or less is a low ridge of
ancient pahoehoe rising from 30 to 50 feet above the surrounding
country. ;
A long narrow belt marked and characterized by many cinder
cones (double semicones built by spatter outfall on both sides of
open rift cracks in parallel linear arrangements) leads upward in a
470 : HARRY 0. WOOD
practically straight line from a point a little west of this conspicuous
group to the summit plateau of Mauna Loa (see the photographs,
Plate VI, aand b). North of this lava ridge and east of the belt of
cinder cones (and to a less extent west of it also) is a comparatively
flat area. For a considerable distance to the northward toward
the summit, the surface of this stands at a lower altitude than the
summit of Puu o Keokeo, and its eastward extension from the
belt of cones, though variable, has a width of from two to three
miles. This makes a very conspicuous upland flat of very slight
grade on a broad dome surface which is itself of very gentle slope.
South of Puu o Keokeo the usual slope of the mountain is resumed.
At the southeast and east this broad, irregular flat passes imper-
ceptibly into the irregular, broken slope of the dome. Several
miles to the north of Puu o Keokeo the flat passes rather quickly,
though the region of transition is indefinite, into the somewhat
steeper slopes which rise to the summit plateau. Viewed from
the ridge south of this flat, Mauna Loa appears as a distinct
mountain until the genetic significance of the long belt of cones is
understood (see the photographs, Plate VI, a and 3).
This belt of cones marks an unmistakable major rift zone,
which joins at the summit with that leading down the northeast
flank of the mountain—a great crust fracture, as a whole slightly
curved and convex to the northwest, through the whole dome of
Loa. This stretches northward across the flat and bounds on the
west that part of it seen on this reconnaissance.
SOURCES OF ERUPTION
As was anticipated, the sources of the eruption of 1916 were
found to lie in this major rift zone. Both the sources of earliest
outbreak, high on the dome, and the sources of flow lie init. Both
were seen to constitute segments of the rift zone, and to be them-
selves rift traces in this zone.
THE SOURCE OF EARLIEST OUTBREAK
A long fissure marked by constant emanation of steam (or
fumes) was seen leading from near the summit plateau of Mauna
Loa down the south-southwest side of the upper slopes of the dome
NOTES ON THE 1916 ERUPTION OF MAUNA LOA 471
nearly to the region where this grades into the great flat—a distance
of several miles (see the photographs, Plate VI, a and 6). For the
most part the steam was clinging to the surface along the line of
fissure; but at one point (and possibly a second) it was rising
definitely in small volume. Unquestionably this eruption marks
a minor rejuvenation in 1916 of the action of rifting through Loa,
and in this upper segment were the orifices, large and small, from
which came the outrush of fumes in the morning of May 19. The
place, or places, where steam appeared to be rising definitely was
well up the slope beyond its transition into the flat. The line of
fissuring marked by steam emanation possibly is interrupted, and
perhaps is offset en echelon. ‘This could not be determined posi-
tively when seen from so considerable a distance. This fissure
leads down in line with the primary system of double semicones
which stretch across the flat along the rift zone from near Puu o
Keokeo. This line of new steam emanation was seen definitely at
all times when the upper slopes were in the field of vision.
THE SOURCE OF FLOW
The source of flow was found to be a freshly opened rift crack
or, more precisely, a long, narrow system of closely spaced parallel
cracks. This ran in a direction slightly oblique to that of the
broad rift zone, but confined well within the limits of it, for some
three miles or more, tending very slightly to the west of north
from Puu o Keokeo across the flat toward the summit. Its upper
limit was not reached on this reconnaissance. : Out of these fresh
cracks gushed the molten lava which streamed away toward lands
in Kona, and toward Kahuku—the streams dividing at the northern
base of the group of old cinder cones at Puu o Keokeo. Only the
Kahuku branches were seen on this reconnaissance.
OBSERVATIONS NEAR THE SOURCE
Our bivouac for the night of May 30-31 was in the lee of an old
triple-peaked, double semicone in the south re-entrant, where its
parts straddle an ancient cinder-choked fissure. This cone was
elongated in the north-south course of this fissure. It had been
the source of an ancient eruption of pahoehoe. This station was
472 HARRY O. WOOD
between two and three miles from Puu o Keokeo in a direction a
trifle east of north. Just west from here, about a quarter of a mile
away, was one of the two larger cinder cones of 1916. This was a
double semicone built on either side of the new rift crack. Both
north and south of this new cone were several other new cones.
A flow of a—a, undoubtedly of the date of 1907, separated us from
the line of vent of the latest activity.
Though greatly diminished, there was still vigorous action at
many points along this line. At first the most active point was a
cinder cone near the northeastern base of Puu o Keokeo, between
two and three miles almost due south of us (see the photograph,
Plate VI, d). Though this was down the wind, which was gentle,
however, explosive coughing sounds could be heard at frequent
irregular intervals; and occasionally red-hot masses were thrown
up into our field of vision. This action continued throughout the
evening and the early part of the night. During this interval the
glow above this vent was considerable, though less than that
ordinarily seen above Halemaumau as viewed from the observatory
at about the same distance (yet, through the disturbed air, it
appeared to be comparable with this). However, at about
8:45 P.M., May 30, a short, sharp earthquake occurred, plainly
felt by all three of us sitting or reclining on the cinders in the
fissure re-entrant of the old cone. (This shock was felt sharply at
Waiohinu and at Kapapala. At Hilea it was felt as the strongest
shock of the entire series connected with this eruption.) Within
less than a minute, but more than thirty seconds, after this shock
there occurred a spasm of greatly increased action at the vent
mentioned, with the jetting of lumps of incandescent lava high
in the air, and a great increase in the glow. However, the action
again quickly subsided to normal. Afterward the action at this
vent declined, at first slowly, but toward 1:00 A.M., May 31, more
rapidly. By 3:00 A.M. the situation of this vent could barely be
made out. When seen again in midafternoon, on May 31, from
near Puu o Keokeo, only a smoking cone appeared. There was no
revival of activity afterward, so doubtless we witnessed the dying
of action at this vent.
In the late afternoon of May 30 a glowing cone was seen,
showing an oven-like orifice, situated at a distance of two hundred
NOTES ON THE 1916 ERUPTION OF MAUNA LOA Alyfe
to three hundred yards to the northeast of the vent just described.
But after night fell the glow above this was negligible.
Between the new cone directly west of our bivouac and the
place of greatest activity just described fumes were rising steadily
from numerous larger and smaller vents and cones. After darkness
came on three of these fuming places exhibited glow. From our
viewpoint this glow came and went intermittently, but this appeared
to be due to drifting fog and fumes alternately concealing and dis-
closing the illuminated fume columns.
The larger new cone just west of our bivouac showed steady,
vigorous glow on the fumes at both its north and south extremities.
These were separated by a dark interval of about 200 feet. No
incandescent matter was thrown from this cone into our field of
vision.
To the northwest, at a distance of about a third of a mile—
hence probably more than 25 miles from Puu o Keokeo in a direc-
tion a little west of north—was a cone which in some respects
exhibited the greatest activity of any of the vents, though it was
not conspicuous for glow. Indeed, it was remarkable for the com-
paratively slight amount of illuminated fumes which appeared to
spread from it. By daylight this cone was not seen clearly, owing
to drifting fog and fumes; but at night it became plainly visible.
It was still building. There were numerous incandescent gashes
and glow-spots on its sides which remained without substantial
change throughout the hours of the night. These were interpreted
as true gashes and orifices in the shell of the cone, throngh which
shone out the incandescent core. Almost incessantly red-hot masses
were thrown out of this cone into the air. Most of these barely
cleared the summit to tumble and roll down the sides of the cone.
The relative motion of these between and among the practically
permanent orifices created the illusion of a steady fountain-play of
fiery particles high above the summit of the cone, rising and falling
like droplets at the top of a jet of water. It required prolonged
observation to correct this impression. At intervals of from
twenty seconds to three to five minutes, larger masses were pro-
jected into the air high above the apex of the cone. These usually
would describe free parabolic curves to fall, apparently, beyond the
base of the cone. In most of these cases no rolling was seen. At
474 HARRY O. WOOD
the time exaggerated estimates were made of the distance of this
cone, and therefore of its height, the height of projection, and the
sizes of the projected masses. Drifting fog and fumes gave a
greatly lengthened perspective effect. Afterward the cone was
visited and found to be distant about one-third of a mile from this
bivouac, and to be about 30 feet in height. Whence the height of
projection of the molten lumps, at highest, was about 45 feet above
the apex of the cone, or 75 feet above its base. No large cinder
lumps were found about it. Though the cone was situated across
the wind from us, a steady, gentle drift of air, loud, staccato,
explosive booms could be heard occasionally accompanying the
projection of the larger masses. This cone was the most spectacu-
lar remaining center of activity of any which came within our range
of vision. On the southwest side of it there was an intermittent
illumination of the thin fumes, of a sort which suggested outflow of
lava coursing away in the southwest direction. However, the later
visit determined that this was a spatter cone higher up the rift
than the head of flow, and that any flowing from it was very local
and confined within a very small area.
Farther north was seen a glowing orifice like an oven which prob-
ably was a gash in a quiet cone. Still a little farther up, rising
fumes were seen, but these showed no illumination at night. After
darkness fell, nine places altogether were distinguished where rising
fumes were illuminated. All of these but one were aligned along
the rift crack northward from the chief vent at the northeast base
of Puu o Keokeo. The other was the oven, situated a little way
to the northeast of this chief vent.
Small quantities of new basaltic pumice, yellow in color, usually
in elongate stringers, were found in close proximity to the rift
source near its head at the east.
THE KAHUKU BRANCHES
The Kahuku branches of the 1916 flow run in southeast and
south-southeast directions from points along the rift source begin-
ning at the base of Puu o Keokeo and extending northward for
more than a mile. On our foot journey, on May 30, from 3:00 P.M.
on we kept encountering little fuming areas lying to the south of
NOTES ON THE 1916 ERUPTION OF MAUNA LOA 475
our route, which indicated the courses of tongues of the new flow.
We approached these quite closely as we neared the rift. The
trunk of this flow departs from the source at Puu o Keokeo in an
east-southeastwardly direction, passing along and around the end
of the ridge of ancient pahoehoe that juts out from the cluster of
cones; there it swerves to the south-southeastward and spreads
down the mountain.
Near the source the lava was pahoehoe in typical surfaces and
in broken crusts and fragments. Except near the source the lava
‘In these branches was a—a wherever they were approached closely
enough for this to be determined. At points on the east these
tongues were thin, from 5 to 10 or 20 feet deep; but at the
south and along the west the lava blocks were piled irregularly
from 20 to 4o feet deep, or high, and were still hot and fuming
on May 31.
In passing from the southernmost point to the eastward and
northward many thin, narrow tongues (from 5 to 8 or 10 feet in
depth, and from 50 to 200 yards in width) were encountered
radiating to the southeast and east. These departed from the
main stream at higher and higher points. Wherever junctions
were seen the departures of these minor branches appeared capri-
cious; that is, no evidences of local damming or pooling were seen.
Though thinner and much less massive than the more western
streams, these were still fuming, and in varying degrees the air
above them was in a state of shimmer from heat. However, the
emanation of the fumes furnished a more reliable indication of
their courses than the heat-disturbed air above them. (Probably
tongues, or ‘‘toes,” project from the main streams on the western
side, as others report who viewed them before they ceased flowing;
we found no opportunity to follow the margin closely and did not
note any conspicuous projections.)
Any adequate cartographic delineation of the complex out-
branching of this Kahuku part of the 1916 flow can be accomplished
only by actual topographic survey. (A reconnaissance survey
was made in June, 1916, by a party under the Hawaii Territory
Survey. An adapted, and in some details corrected, modification
of this follows, as Fig. 1. A general conception of a long, narrow,
476 HARRY O. WOOD
branching flow (about 53 miles long) to the southeast and south-
southeast, with many minor spreading tongues, is all that could be
developed as a result of this reconnaissance.
When we were in the neighborhood of the new lava, smells of
subliming sulphur, sulphur acids, charcoal, and cinders were very
noticeable. There were also smoke smells from burning vegetation.
Gi
Cy
5
Pel Crone EOWEO
we
-
<< (1868) ® 12,000
\ 1,50p
S
\ Ls ji
ans
KT
LUDA
Lp
Rar.
L/ pa 3 He Woeien
“Uo y
oS: “Lays y Sed Wy; ni
Scace : MILes.
10 5
SS
Fic. 1.—A diagrammatic map of the 1916 flow and the neighboring region,
adapted from a reconnaissance survey by the Hawaii Territory Survey, June, 1916,
with corrections and modifications suggested by field work and reports. This plate
also shows a diagrammatic correction of the course of the 1907 flow near Puu o Keokeo.
IV
The thorough exploration of the region of the source of flow
made in company with Dr. A. L. Day in late June and early July
served to correct or confirm the findings of the hurried reconnais-
sance made in the last of May, and to enlarge their scope materially.
-Also many photographs were made, affording a fairly complete
pictorial record of the results of the action at the source. However,
NOTES ON THE 1916 ERUPTION OF MAUNA LOA 477
our efforts on this occasion were devoted in large measure to
observation of details and to the examination of evidence bearing
on the physical-chemical conditions and the mechanisms of flowing
in the lava streams near their head. ‘The ideas considered in this
connection are best left for future discussion by the writer’s com-
panion on this expedition. Here it will suffice to say that much
was seen tending to confirm, and some things tending to modify,
the writer’s conception of the mode of flow of a—a, as exempli-
fied by the action observed at the front of the Honomalino
branch, described above. Consequently, the space here devoted
to this more thorough work is small in proportion to its relative
importance.
On June 27, 1916, we set out from the observatory and went by
motor to Honomalino, and thence with horses up the southeast
flank of Loa to a point in Kahuku, above Papa, at an elevation of
about 6,500 feet above sea-level. Here we made camp on barren
ground a little above tree-line on this part of the mountain, close
beside a short narrow branch of the new flow—the most north-.
western of all its definite branches. The six days, June 28—July 3,
we spent in exploration of the source region. On July 4 we returned
on horseback to Honomalino and by motor to the observatory.
This camp site was situated on the regular, gentle slope of the
mountain dome, between 14 and 2 miles below the junction of this
branch of flow with the rift source. Everywhere here the old sur-
face was of ancient, rusty-red pahoehoe and a—a commingled in a
complicated pattern—except for an area of gray pahoehoe, younger,
but still very old, found about 1; miles above camp. Nearly all
our way upward from camp to the source led alongside the new flow
over slopes below the limits of the great flat above Puu o Keokeo,
but as the 1916 source was closely approached these slopes graded
into this upland plain west of the new rift cracks.
The sources of all the branches of the 1916 flow lie in the new
rift segment. The ror5 rift here is a newly developed fissure, or
in most places a very narrow system of closely spaced fissures (the
primary group together nowhere more than 30 feet wide and nearly
everywhere much narrower) which extends from the northern base
of Puu o Keokeo for a distance estimated closely at 33 miles in a
478 HARRY O. WOOD
direction almost exactly magnetic north’ (see Plate VI, a). This
fissure, or system of fissures, is practically uninterrupted and, —
except for a short curving segment at the north end (see Plate VI, c),
it follows a straight line. It lies wholly within the limits of the
great rift zone, but its course is slightly oblique to the general
trend of that, which is about N.N.E. (see Plate VI, 6), and this
suggests a major shear through the mountain. It ends on the
north at an old red cinder cone at an elevation of about 7,480 feet
above sea-level, while at the south its point of interception with
Puu o Keokeo is at an elevation of about 6,600 feet (see maps,
Plate I and Fig. 1).
Besides this chief rectilinear fissure, or primary group of fissures,
there are a great many secondary cracks, running roughly parallel
with the system, especially on the west; on the east there are few.
Many of these opened after the chief outpouring of lava was over,
for they traverse the fresh flows. (It is notable, moreover, that
earthquakes continued to increase in number and energy until after
the eruption began to decline definitely.) Many, however, traverse
the older surface neighboring the source; and here they appear as
consistent extended fissures in the more or less solid basalt of the
mountain, but also there is noted a tendency for the fissuring to be
continued from one crack to another through offsets en echelon (see
Fig. 2).
Though miniature dislocations have resulted necessarily, there
is no observable tendency to any general vertical dislocation; but
the repeated evidences of offsets en echelon strongly suggest a general
horizontal shear. There are, however, no sufficiently well-indicated
and extended landmarks, surface features, or structure lines to
afford a real test or proof of this. Where these cracks traverse the
old surface and where they cut the new flow, there generally is no
evidence of gas outrush, bulging, or fumarolic action, or any evi-
dence of heat emission.
Altogether, as study progressed the conviction gained force
that the amount of rending of the mountain dome here seems out
« The general magnetic declination in Hawaii is about N. 10° E., but large local
variations make the application of corrections so uncertain that the direct magnetic
reading is given here by preference.
NOTES ON THE 1916 ERUPTION OF MAUNA LOA 479
of all proportion, either to the outrush of imprisoned gas from this
vent (for gas emission throughout the flow was apparently small in
volume and relatively very quiet), or to the momentum-pressure
of the outpoured magma. Also, the features of the rifting and
their distribution appear to differ from those that should be expected
to result from such causes. (Resemblances of this action to fissure-
eruption phenomena in Iceland are noted below. See especially
Plate VI, c, and Figs. 2 and 3.) Moreover, as just mentioned,
great numbers of weak to moderately strong local earthquakes
were registered at the observatory 30-35 miles from the source of
action, the energy of these zmcreasing even after the cessation of
forward movements of the flow, but while the vents at the source
Fic. 2.—A panoramic view looking N.N.E., showing the old red cinder cone
riven by new cracks, and the solfatara at the head of the new rift (the figures of the
men give scale); and black lava, of 1907, at the base of the cone, which is a little
over too feet high.
were still freely open. Several of these were felt definitely over a
considerable area, having a radius of much more than 30 miles;
and one or two of these were quite sharp at the observatory. The
question of their origin will be discussed in a later paper.
In brief, for many reasons the conception that this eruption
was, in part at any rate, primarily a tectonic event must be examined
thoroughiy and not put lightly aside. No very violent action was
observed or suggested, nor any such as would be expectable were a
rift like this to be produced by explosive forces, or by upthrust
from a confined substance tending to expand rapidly or seeking
immediate outlet to the surface—as volcanic potential usually is
480 HARRY O. WOOD
hypothesized. The lips of the rift were not outforced or uplifted,
nor were radial cracks produced. Such action of lava and gas out-
rush as was observed would be expectable if eruption were permitted
through tectonic rending of the mountain shell in a manifestly
weak zone, thus unsettling the physical-chemical equilibrium of the
gas-charged magma afforded exit to the surface and the atmosphere
in this way. Many of the eruptions of Mauna Loa have exhibited
similar peculiarities, perhaps most of those observed. This whole
aspect of eruption here deserves a much more thorough discussion
than can be given it in this paper. However, to the mind of the
writer, the phenomena of the 1916 eruption appear to illustrate
admirably the conception so clearly stated by Geikie,’ if only the
expression ‘‘tectonic strain’ be substituted for his phrase “terres-
trial contraction,” with the emphasis in this instance on tectonic
strain as a cause.
The fact that some of the circumstances of this eruption might
be interpreted adversely to this view may be discussed more
advantageously in a systematic study of the seismic accompaniment.
Altogether this conception deserves careful attention. Of course, it
must not be considered to invalidate or displace views developed
from, and applied to, other modes of eruption in other lands, but it
must not be rejected simply because it is different from them.
Moreover, there is no disposition to overlook the very real applica-
tion of more commonly recognized modes of eruption in this recent
outbreak, or in other eruptions in Hawaii. It is intended merely
to give this tectonic mechanism emphasis and to point out its
possible, or probable, local predominance.
Lava did not well out of the new rift along its whole length.
Hence we may subdivide it, recognizing two segments: a longer
flow-source segment, and a shorter solfatara-spatter-cone segment
stretching on up the mountain beyond the head of flow.
THE FLOW-SOURCE SEGMENT
For about 23 miles from Puu o Keokeo in a direction about
N. 3° W. mag. the new rift is indicated by a system of open fissures
straddled by seven double semicones, varying from 20 to too feet
t Ancient Volcanoes of Great Britain, I, 10-13.
NOTES ON THE 1916 ERUPTION OF MAUNA LOA 481
in height and from 50 to 200, or more, feet in the length of the
greater diameter, built of pumice, cinders, and spatter outfall—
these are also primary sources of flowing streams—with many
smaller cones and mouths intervening. ‘The resemblances of these
cones and fissures, and their interrelationships, to features in
Iceland—especially along the great Laki fissure—as described by
Geikie? and those from whom he drew, and as exhibited in the views
by Anderson reproduced by him, is very striking indeed; few would
Fic. 3.—A view looking north into the gash of the second largest cone of 1916,
situated a little south of the head of flow. Its character as a double semicone, built
of ejected products, and the channel of outflow from the gash are shown. This cone
is about 70 feet high.
question that the course of action was similar in both regions
(see the photograph, Fig. 3, especially). Moreover, at prac-
tically all points along this segment lava welled out and flowed on
both sides of the open rift in southeast, south, or southwest direc-
tions down along the course of the fissure system and out along
narrow, jutting tongues of greater or less length, as well as down
the greater streams. One of these, the largest, led toward Kahuku
(see the photograph, Plate VI, d), and three others led toward
1 Op. cit., II, 260-65, Figs. 292, 293.
482 HARRY O. WOOD
Kona, two of which were relatively small, though not to be mis-
taken for tongues. Of all these greater streams the Honomalino
branch was longest (see the map, Fig. 1).
At its upper end the area of the new flow is narrow, varying in
width up to a third of a mile, and elongate parallel to the rift; its
margins here are irregular and lobate on a small scale.
THE SOLFATARA-SPATTER-CONE SEGMENT
For a mile above the head of outflow the open-fissure system
continues, at first in a direction about N. 3° W. mag., but in its
last third it curves gently eastward, so that the direction from the
foot of the flow-source segment to ‘the head of this upper segment
is N. 2° W.mag. It is marked by an almost continuous line of fresh
solfataric action along which, in numerous protected niches, very
delicate, feathery, sulphur crystals were subliming in considerable
quantity, apparently in unusually pure aggregates (see the photo-
graphs, Plate VI, b,c). This action, and also all conspicuous
fissuring, ended in the flanks of an old red cinder cone, whose
summit is about 7,480 feet above sea-level and its base 7,375+,
which stood directly in the course of the major rift belt not far
from its western margin.
Also there were observed along this segment three new spatter
cones, the largest 30 to 4o feet in height, wholly isolated from the
area of continuous flow; and several small spatter mouths, one of
which was situated very near the upper end of the segment at an
altitude of 7,370 feet above sea-level.
Near their head the new flows are thin pahoehoe, either in
smooth sheets traversed by rift cracks of later origin, or, more
commonly, broken and torn crusts of pahoehoe transported and
piled into an irregular and confused surface (see the photographs,
Plate VI, d, and Fig. 4). Near the edges of flow, and the
edges of festooned flow channels, rough, a—a-like textures are seen
in all intermediate phases between “pulled”? pahoehoe and
cindery a—-a. Down their courses the flows become thicker and
their surfaces more irregular and fragmented, passing finally,
within a mile or two, through slaggy phases, into unmistakable
NOTES ON THE 1916 ERUPTION OF MAUNA LOA 483
cindery a—a. These flows spread indiscriminately over the various
materials of the old surface and near-by areas of the lava of 1907.
At and near the sources the following types of new lava were
observed:
a) PRODUCTS OF FLOW
Pahoehoe, which here exhibits various surface textures deter-
mined by the interrelationships of different conditions during the
progress of flowing, such as the degree of viscosity (dependent in
part upon the temperature, the gas content, its state of solution,
Fic. 4.—A view looking southeast, showing the cone at the head of flow, thin,
fresh pahoehoe spread over old pahoehoe and a—a, and the south terminal of the line
of solfataras. This cone is from 15 to 20 feet high.
the amount of the crystalline content, etc.), the mass, and the
gradient of the surface, all these influencing the rate and manner
of flow and the consequent action of subsurface traction on the
crusts.
Textures
Pumiceous texture, a finely vesiculated spongy surface very
closely resembling basaltic pumice in color and structure.
Lacy texture, a more coarsely vesiculated spongy surface modified
by flow tractions so as to resemble patterns of complicated lacework.
484 HARRY O. WOOD
Ordinary textures, comprising surfaces of varying character and
degree of vesicularity, and surface vesicle patterns, difficult to
illustrate or describe except at great length, but common in all
fields of pahoehoe.
“Pulled” texture, seen in incipiency in the lacy texture, but
extended to most of the ordinary types wherever, through con-
tinued traction, the surface was greatly sheared after it had stiffened
or partially set. This gave rise to stippled and bladed surfaces,
and to actually fragmented surfaces, so thus, by degrees, it passed
over into a slaggy a—a texture.
The textures most prevalent in the source region were the
pumiceous texture and the ‘“‘pulled” texture.
Slag, a product of flow made up of rough fragments varying in
size and in character from torn and wrapped pahoehoe crusts to
cindery a—a lumps.
In a general way, here slag was characteristic of the flow
channels near the source and of an intermediate region down the
flows between the typical pahoehoe and a-a stages.
A-—a, rough-surfaced block lava, typical of by far the greater
part of all the 1916 branches of flow.
b) EJECTED PRODUCTS
Basaltic pumice in three distinguishable varieties which, of
course, grade into each other: (1) a very finely vesicled variety of
light-yellow color, almost a thread-lace scoria in structure; (2)
ordinary yellow to brown basaltic pumice, with fused surfaces,
resembling pulled molasses candy; and (3) a more coarsely vesicu-
lated brown to black pumice which grades with the increasing size
and the decreasing numbers of vesicles into
Cinder lumps, which, in turn, grade with decreasing visstenl ial
and increasing density into
Slag lumps, (1) some of which exhibit a surface like obsidian;
(2) others a surface like a-a.
All these were observed all along the rift and about the spatter
cones, but the pumice phases were very abundant near the south
NOTES ON THE 1916 ERUPTION OF MAUNA LOA 485
end and around the major cones (see the photograph, Plate VI, d),
while slag lumps were more characteristically found about the
spatter cones and along the northern part of the rift course.
Also there was suggestion that the ejection of slag lumps con-
tinued after the action of pumice ejection was over, but no proof
of this could be elicited.
DISTANT OBSERVATION OF THE UPPER SOURCE
The upper source of the 1916 eruption, where the outbursts of
fumes occurred on May 109, has not yet been visited, so far as the
writer is aware. ‘This is a place on the great mountain dome very
remote from trails, and travel with horses over the barren, untracked
lava is both difficult and dangerous. This source lay between
13 and 15 miles from our camp—a distance altogether too great to
cover on foot in the short time at our disposal over going so rough
as that prevailing within and along the rift belt.
However, this region could be seen plainly through the clear
mountain air from the old cone at the head of the lower rift, at a
distance of a little over 9 miles. With binoculars magnifying
eight diameters some of its characteristics could be made out. It
lay within, and was much elongated parallel to, the axis of the
major rift zone. It extended from near the edge of the summit
plateau down the somewhat steep upper slope nearly to the great
flat which les north of Puu o Keokeo. It appeared to have a
moderate breadth, say a quarter of a mile. Fumes were clinging,
or slowly rising, along its axis, probably from a system of fissures
(see the photographs, Plate VI, a and 0). In two places, appar-
ently not far apart, about two-thirds of the way up its course,
definite columns of rising fumes could be made out frequently.
On one occasion, in the late forenoon of July 3, one of these columns
was estimated to reach upward 500 feet before it spread out and
dissipated.
At all times when it was clearly visible the surface of this source
area, on both sides of the line of fumes, appeared of very light
color, even whitish. It could not be determined whether this was
due to the sheen of new pahoehoe, to efflorescence of sulphur or
486 HARRY O. WOOD
sulphur salts, or to the escape of fumes in very small quantity
from numerous cracks distributed over the area. No other expla-
nations of this light color suggested themselves.
Cones of medium size could be seen within the area. Fumes
were rising from them and around them. | It is uncertain whether
these were new cones, or old cones freshly riven. In general,
here, eruption does not take place through old fissures reopened;
but exceptions are known. There was no great amount of action
at night at this upper source. However, the fuming action in the
cleft and on the sides of one of these cones strongly suggested that
it was of new origin.
Certain features of older origin, found in the neighborhood of
the lower source, claim bare mention here.
Above Puu o Keokeo there is a long, narrow flow of fresh,
black a—a (with long narrow tongues projecting from it over the
great flat to the eastward), which stretches along the eastern
margin of the 1916 flow source, and in part lies under the 1916
outflow. This begins much farther up the mountain, in the course
of the rift, than the 1916 head. Undoubtedly this is lava of 1907,
and it is mapped by Baldwin as of this date (see the map, Plate I).
However, on his map it is shown as extending down past Puu o
Keokeo on the east of that group of cones. Nevertheless, on the
writer’s short reconnaissance he went on horseback up between the
flow of 1887 and the Kahuku branch of 1916 onto the ancient
lava ridge that projects uninterruptedly eastward from Puu o
Keokeo without crossing this lava stream. Hence it is clear that
no part of this flow follows the course past Puu o Keokeo shown on
that map. This upper stream did, however, pass down on the
western side of Puu o Keokeo. While the detailed expression of its
course, therefore (and of that of a contributory stream from a source
below Puu o Keokeo), must await adequate topographic survey,
the writer has diagrammatically sketched its course on the west of
Puu o Keokeo on the map, Fig. 1, showing these southwestern
flows. However, he has not attempted to indicate the long tongues
which project southeastwardly over the great flat north of Puu o
Keokeo, on account of want of data, and of the cartographic con-
fusion that might result.
NOTES ON THE 1916 ERUPTION OF MAUNA LOA 487
The 1916 eruption was of small magnitude compared with
earlier action originating near its source. ‘There are accumulations
of ancient pumice far greater in depth and spread than that ejected
at this time. The older double semicones round about are higher
and greater and their gashes and fissure systems on a larger scale.
One such ancient source—whimsically designated in field notes
as “the lunar crater’’ because of a steep and relatively high peak of
riven blocks which stood near the center of its large, circular
depression, or crater-like area—had been the spring of a lava flood
vast in proportion to the recent flow. All about this old depres-
sion, except for its gap at the south, was a high rampart built of
huge cinder blocks piled confusedly, and outward from the top of
this a slope built of small cinders and pumice fragments fell away
gradually.
Also the cones at Puu o Keokeo point to action of far greater
magnitude at the time of their building than recently—greater
than any action of historic date on the south flank of the mountain.
These cones, however, are only a conspicuous group in the well-
marked belt extending from the summit down the slope below
them. ‘Though perhaps these are the largest of all, there are others
of comparable size, both above and below. The suggestion, there-
fore, attributed to the late S. E. Bishop, that Puu o Keokeo is a
vent distinct from Mauna Loa, but subordinate to it, probably
must be dismissed. ‘This point, though a digression, is interesting
and important, since it might be considered to bear on the question
of the genesis of the 1916 eruption.
In connection with this same point it is worthy of note that the
eruptions of 1868 and 1887 (and probably of 1907 also) were pre-
ceded by outbreaks of fumes and lava much higher up the moun-
tain than their eventual heads of flow—action similar to that
preceding the 1916 flow.
In 1868 such action broke out in the evening of March 26, with
a further outburst in the early morning of March 27 “‘a little to the
southwest of the summit,” followed by outflow from low sources,
at the southwest on Kilauea on April 2, and at the south-southwest
on Mauna Loa on April 7. This upper outbreak was very near
the summit.
488 HARRY O. WOOD
In 1887 the first outbreak was in the evening of January 16 at
an elevation of about 11,500 feet on the southwest, and flow began,
from a little below Puu o Keokeo, in the evening of January 18.
The exact places of. these upper outbreaks were not mapped in
either instance.
In 1907, on account of weather conditions adverse for distant
seeing, only vague accounts were given. Nevertheless, mention is
made, perhaps doubtfully, of an outbreak judged to be at the
summit, preceding by a few hours the outbreak of flow lower on
the flank.
Hence, though the places of outflow in several of these eruptions
have been found suggestively near to Puu o Keokeo, still there is
no doubt that the eruptive action in all cases extended far up the
south flank of Mauna Loa beyond this group of old cones. More-
over, critical study of the distribution of these heads of flow develops
no causal association with this as a center.
JouRNAL oF GEoLocy, VoL. XXV, No. 5 PLATE VI
a b
pe ee : aaa ee Bo ~ es sc = asec - i i
G d
a, a view looking a little to the north of east from Puu o Keokeo, showing the flow source. The new
rift line, indicated by arrows, traverses the middle ground obliquely. Note the gashed cones; also note
the faint streak of fumes, line 3, near the mountain summit, 14-15 miles away—the source of the fume
outburst of May 19. Line r indicates the larger cone at the chief head of the Honomalino stream, Line 2
indicates the largest 1916 cone at the chief head of the Kahuku branches.
b, a view from an old cone a little N.W. of the head of flow looking about N.E. at the northern portion
of the new rift, indicated by short arrows at margin, marked by solfataric action. Note also the streak of
fumes, and the light-colored area, at the source of the outbreak of May 19, line 3.
c, a view looking south toward Puu o Keokeo from the old cone at the head of the new rift. Note the
slight curve in the line marked by solfataric action, with new cones beyond; line 1 indicates one of the
larger new cones at the chief head of the Honomalino stream, line 5 indicates the small cone at the head of
flow, and the lines 4 indicate Puu o Keokeo, 33 miles away. In the foreground is a freshly riven spur of
the old cone.
d, a view from near the eastern edge of the new flow, looking S.S.W.—a detail of the source near its
southern end, showing the cone at the head of the Kahuku branches; line 2, a short tongue of 1916 lava
projecting eastward, and old surface in the foreground. A considerable fall of new, basaltic pumice partly
covers both old and new lava here. The lines 4 indicate Puu o Keokeo.
A PROPOSED DIP PROTRACTOR
CHESTER K. WENTWORTH
University of Chicago
The device herein described consists of a celluloid chart giving
the dip of any plane along a line at any given angle with the strike.
Several tables for this purpose have appeared which, for office work,
are entirely satisfactory. The protractor has in a certain class of
ae co
ee OOS GN ES 0.000 AP
ah SAO aan ae
—
DOS
Fig 1.
fieldwork, which will be outlined below, a superiority over the
tables and has not, so far as I am aware, been before described.
The protractor consists of a rectangular plate of transparent
celluloid ruled and numbered as shown in Fig. 1. The circular
curves represent each a given angle of dip of the plane, i.e., the
maximum angle commonly denoted dip. The intersection of
the appropriate curve with the radiating line of direction of the
required dip cuts off an ordinate, or distance from the horizontal
489
490 CHESTER K. WENTWORTH
diameter line, the value of which in terms of the vertical scale at
the side is the required angle of dip.
As an example, let it be required to find the angle of dip in a
direction 4° from the strike, of a plane of which the maximum
dip is 5°. Following the 5° curve to its intersection with the 40°
radial line and interpolating this point between the 3° and 4°
parallel lines we find the value of approximately 3°15’ which is the
desired angle of dip.
The particular field of usefulness of this device is in projecting
the plane of a given stratum whose dip and strike are known from
a plane table set up on the outcrop. We have in this case the dip
and strike of the plane and the direction of the line of sight for
which we wish the dip. The horizontal diameter line of the pro-
tractor is laid parallel to the strike as recorded on the oriented
plane-table sheet and the alidade set on the required line of sight
with its edge passing through the center on the protractor. At the
intersection of the ruler edge with the appropriate dip-curve is read
the required angle of elevation or depression to be set on the tele-
scope to project the plane. This graphic solution of the problem
in the field and directly on the plane table is much more rapid
than a combined protractor and dip-table solution and is suffi-
ciently accurate for reconnaissance and mapping purposes where
projecting the outcrop on topography is the “‘best guess”’ the field
man has in many instances.
The protractor as shown is made only for angles up to 10°
because occasions for projecting dip in cases of higher dips are
much more rare and correspondingly less accurate, and greater
accuracy for the low angles is attained by putting fewer lines on
the celluloid. The radii of the several dip circles are constructed
proportional to the tangents of the respective dip angles, as is also
the spacing of the parallel horizontal lines which are tangent to
them. Thus if A is the nominal dip of any plane, B the required
oblique dip, and C the angle of obliquity, we have
tan B
Sin (C=
tan A
A PROPOSED DIP PROTRACTOR 491
and a circular arc may be used for the dip-curve. For the low
angles as shown on the protractor in Fig. 1 the tangents are so
nearly proportional to the angles that the spacing intervals depart
from equality by an inappreciable amount only. There should be
no difficulty in reading required dip angles to the nearest two or
three minutes of arc with the protractor as shown, and several
protractors might provide for the whole range up to ninety degrees
for those requiring the higher angles.
PETROLOGICAL ABSTRACTS AND REVIEWS
ALBERT JOHANNSEN
ABENDANON, E. C. Considérations sur la composition chimique et
minéralogique des roches éruptives, leur classification et leur
nomenclature. La Haye, 1913. Pp. 34.
ANDERSEN, OLAF. ‘‘The System Anorthite-Forsterite-Silica,”’
Amer. Jour. Sci., XX XTX (1915), 407-54.
A description and discussion of experimental methods used, and
results obtained, by fusing quartz, alumina, calcium carbonate, and
magnesia. Solid phases of anorthite, forsterite, cristobalite, tridymite,
clino-enstatite, and spinel were observed, and their thermal and optical
properties determined. Anorthite and silica form a simple eutectic
system, forsterite and silica form a system with an unstable compound,
while anorthite and forsterite form no true binary system. The appli-
cation of the results to igneous rocks is pointed out.
ArscHinow, W. W. On Inclusions of Anthraxolite (Anthracite)
in Igneous Rocks of Crimea. Petrographical Institute “ Litho-
gaea,” Publication No. 4. Moscow, 1914. Pp. 15. (In
Russian language.)
The term anthraxolite, originally proposed by Chapman for anthra-
cite found associated with quartz and pyrite in certain veins in the Lake
Superior region, is used by Arschinow for all bituminous substances.
Such substances occur in the form of small, black inclusions in igneous
rocks in two places on the southern shore of the Crimea. Since it also
occurs as vein and cavity fillings in the rock, and in many cases is asso-
ciated with calcite and quartz, it was probably formed, after the cooling
of the magma, by the destructive distillation of bituminous substances
disseminated in the stratified rocks.
Batt, SyDNEY H., and SHALER, Mitrarp K. . “Contribution
a l’étude géologique de la partie centrale du Congo belge y
compris la région du Kasai,” Ann. soc. géol. Belgique, 1913,
199-247, map 1, pl. 3.
The writers found diabase, granite, diorite-gneiss, granitoid-gneiss,
chlorite-schist, and amphibole-gneiss in this region.
402
PETROLOGICAL ABSTRACTS AND REVIEWS 493
Beck, Kurt. ‘Petrographisch-geologische Untersuchung des
Salzgebirges an der oberen Aller im Vergleich mit dem Stass-
furter und Hannoverschen Lagerstattentypus,” Zeztschr. f.
prakt. Geol., 1911, pp. 23, figs. 6, pl. 1.
Discusses the sequence of deposition and the characters of certain
salt deposits, and compares them with the well-known deposits in Stass-
furt and Hannover.
Beck, Kurt. ‘Petrographisch-geologische Untersuchung des
Salzgebirges im Werra-Fulda-Gebiet der deutschen Kalisalz-
lagerstatten,” Zeitschr. f. prakt. Geol., XX (1912), 133-58,
figs. 12, pls. 2.
Discusses the stratigraphical and structural relationships of the salt
deposits of Werra and Fulda.
BrecER, P. J. “Zinnerzpneumatolyse und verwandte Erschein-
ungen im Kontakthofe des Lausitzer Granits,’’ Neues Jahrb.
Min., Geol., und Pal., 1914 (II), 145-82, figs. 4, pls. 2.
The writer discusses the contact zone of the Lausitz granite, describes
various pegmatites, greisens, etc., and concludes that the chlorite in the
highly metamorphosed greywacke is a pneumatolytic mineral due to the
adjacent granite.
Benson, W. N. “The Geology and Petrology of the Great Ser-
pentine Belt of New South Wales” (in five parts), Proc.
Linnean Soc. New South Wales, XXXVIII (1913), 490-517,
569-96, 662-724; XL (1915), 121-73, 540-624; maps 6,
figs. 29, pls. Io.
The region described in these papers passes north and south through
Tamworth, and lies between Sydney and Brisbane, about 280 miles
from the former. Through the center of the region extends a north-
and-south fault, separating it into two portions. Along this fault is
a series of intrusions of rocks now serpentine, not continuous but form-
ing separate patches from too yards to 30 miles in length and from a few
inches to nearly two miles in width. The formations indicate that
after a long period of sedimentation heavy orogenic pressure came from
the east, folding and metamorphosing the eastern series, but having
less effect on the western. The pressure caused the formation of an
overthrust fault which became the channel for the ascent of certain
494 PETROLOGICAL ABSTRACTS AND REVIEWS
basic rocks. The sedimentary formations, which prior to the folding
had a thickness of over 30,000 feet, are described, and a list is given of
the fossils found.
The igneous history begins in Lower Devonian times with the
extrusion of spilitic lavas and tuffs. In the Middle Devonian similar
rocks attain a thickness of over 7,000 feet, and associated with them are
2,000 to 3,000 feet of intrusive dolerite, in many cases albitized. In
Upper Devonian times some 3,000 feet of agglomerates were formed,
and, after a long period of quiet, rhyolites, andesites, and tuffs appeared
‘in the Lower Carboniferous. The intrusion of the peridotites then
followed, chiefly along the fault line previously mentioned, and probably
during the crust-movement at the close of the Carboniferous. Gabbros
and eucrites came later than the serpentine, and cutting these are dikes
of dolerite. From latest Carboniferous to early Mesozoic times came
a long series of granitic intrusions consisting of granodiorites and por-
phyries, and titanite-, tourmaline-, and other granites. Besides these
rocks there are numerous lamprophyres whose time period was not
determined. Following the Permo-Carboniferous was an era of great
crumpling, then followed a long period of erosion which exposed the
granite, and then a later period of sedimentation. During Tertiary
times the formations were largely volcanic, and thick flows of basalt
occurred. A great period of elevation and block-faulting closed the
Tertiary.
The second paper deals with the geology of the Nundle District,
and the third with the petrology of the entire region. Various rocks are
described: spilites, used in the sense of Dewey and Flett for lavas with
sodic feldspars, contain acid oligoclase and augite with some secondary
chlorite and epidote and with or without magnetite; the so-called
“‘keratophyres”’ are composed almost entirely of acid oligoclase with some
interstitial chlorite from augite; the dolerites consist of plagioclase
(andesine to albite), augite, magnetite, with a little quartz and various
accessories, and are medium-grained. The term dolerite is apparently
used in a different sense from that common in the United States,
where it signifies a coarse-grained basalt containing a basic plagioclase.
The writer speaks of albitization proceeding inward in the feldspars,
by which he means, apparently, that the sodic rims are secondary. It
would seem more probable that the zonal rims are primary. The rock
thus appears to be an augite-andesite. The peridotites are chiefly
harzburgites, but there are local occurrences of dunite and lherzolite.
With the absence of olivine, enstatolites occur. Associated with the
PETROLOGICAL ABSTRACTS AND REVIEWS 495
peridotites and pyroxenites are rarely amphibolites and olivine-gabbros,
more commonly eucrites and anorthosites. The silicic rocks described
are felsites, granodiorites, and various granites. Malchites, granite-
porphyries and quartz-porphyries, and minettes, vogesites, and camp-
tonites occur as dikes.
Twenty-two analyses are given of rocks of this region, unfortu-
nately showing a number of typographical errors and errors in proof-
reading owing to the absence of the author from the state during the
passage of the paper through the press.
The fourth paper deals with the dolerites, spilites, and keratophyres
of the Nundle District. In this the statement is made that the feldspar
of the keratophyre is pure albite and not acid oligoclase, as stated in the
earlier paper. The spilites, dolerites, and keratophyres all appear to be
-intrusives in the sediments. Certain of the spilites, in the opinion of
the author, were intruded into soft mud, for they show a pillow structure.
Most of the rocks are rich in primary albite.
The fifth paper is on the geology of the Tamworth District. A series
of radiolarian claystones was deposited at shallow depths on a steadily
sinking sea-floor in Devonian times. There were two periods of volcanic
activity, and masses of tuffs and agglomerates accumulated in the sea,
and spilites, dolerites, and keratophyres were intruded. The total thick-
ness of the series is unknown, but apparently about 12,000 feet are
Middle and Upper Devonian. Faulting and folding took place in the
Carboniferous period, followed by peridotitic intrusions, and later by
intrusions of granite. A final eruption of basalt occurred in the Ter-
tiary period.
BERKEY, CHARLES P. “Petrographic Range of Road-Building
Materials,” School of Mines Quart., XXXV (1913), No. 1,
pp. 6.
BLANCHARD, RALPH C. The Geology of the Western Buckskin
Mountains. Dissertation. Columbia Univ., 1913. Pp. 80,
numerous figs.
BoEkE, H. E. “Bemerkung tiber die Theorie von J. Johnston
beziiglich des Verhaltens fester Stoffe unter ungleichformigem
Druck,” Cenitralbl. f. Min., Geol., u. Pal., 1913, 321-24.
496 PETROLOGICAL ABSTRACTS AND REVIEWS
BoEKE, H. E. ‘‘Die Methoden zur Untersuchung des molekular
Zustandes von Silikatschmelzen,’ Neues Jahrb. Min., Geol.,
u. Pal., B.B., XX XIX (1914), 64-78.
BoEkE, H. E. Grundlagen der phystkalisch-chemischen Petro-
graphie. Gebriider Borntraeger, Berlin, 1915. Pp. xii+-428,
figs. 168, pls. 2.
In view of the important bearing of recent physico-chemical research
upon the problems of the origin of igneous rocks, this work is most timely
and acceptable, and is to be recommended to all advanced students of
petrology. It is an invaluable summary of work already done, and
contains many suggestions for future work.
The author not only presents the results of previous work but
describes in detail the methods and apparatus by which these results
were obtained. The subject is presented in a very clear and orderly
manner, and as simply as is compatible with the nature of the subject.
In his treatment the author follows the inductive or synthetic method,
that is, he describes first the behavior of the simplest constituents of
rocks under known conditions of composition, temperature, pressure,
time, etc., and compares the results with those found in nature. Begin-
ning with the magma, he traces it through all stages of cooling and
through the gradual changes which take place in its solidification products.
In older petrologic textbooks there is a great variety of views as to
observed phenomena, primarily because so many factors must be taken
into consideration, and one or another may predominate. Since in
many cases there are more unknown than known factors, a single solu-
tion may be impossible. The synthetic method seeks to determine, by
exact and systematic investigation, the action of each factor, such as
temperature, pressure, capillarity, etc.
In most provinces of petrology only the beginning of inductive
research has been made, and the experimental work so far is no more
than a groping after the truth. Little can be said at the present time
as to the formation of rocks from their complicated magmas, and physi-
cal chemistry can not yet settle such questions as the origin of mag-
matic differentiation, the relation between the alkali and alkali-lime
rocks, and that of dike satellites to parent rock, etc., on account of the
lack of reliable data.
The present text is so comprehensive that only the very briefest
outline of the contents of the various chapters can be given; a list of the
PETROLOGICAL ABSTRACTS AND REVIEWS 497
subheads alone would require five or six pages of this Journal. After
-a short discussion of homogeneous and heterogeneous equilibria, the
subject of magmatic rock-formation is taken up. Under this head the
author treats of the essential components of magmas and the melting-
points of minerals, and includes a discussion of various methods of
determining melting-points and of obtaining and measuring high tempera-
tures in the laboratory. He speaks of the alteration of melting-points
at different pressures, of uniform and non-uniform pressures, and of
overheating and undercooling. Under the properties of silicate melts
are included internal friction and diffusion, surface-tension, electrical
conductivity, influence of gravity and centrifugal force upon the com-
position of melts, measurements of density at high temperatures, etc.
Following a chapter on the inversion points of minerals, he discusses
the genetic significance of melting- and inversion-points in rock-forming
minerals. He describes two, three, four, etc., point systems, and then
devotes about seventy-five pages to the physico-chemical, especially
the thermal, properties of the more important rock-forming minerals.
In the second division of the book the author takes up the gases
in magmas—their nature, their solubility in melts, and their equilibrium.
In the third division he treats of the pegmatitic, pyrohydatogenic, and
hydrothermal phases of the solidification of magmas. Here are included
the properties of water at high temperatures, and a general discussion
of the formation of minerals in systems with volatile components.
Then follows a chapter on the synthesis of pneumatolitic and hydato-
genic minerals; then hydrothermal synthesis, solubility of the common
products of hydrothermal mineral-formations, alteration with tempera-
ture and pressure of salts which are but slightly soluble, the relation-
ship of solubility and size of grain, succession and paragenesis of the
hydrothermal ores and vein-minerals, zeolites, etc.
The fourth division is devoted to weathering or colloid mineralogy;
the fifth to sediments. Here is included a long discussion of salt deposits
(43 pp.) to which the author has devoted considerable study. The book
closes with a short chapter on metamorphism and a double column index
of 26 pages.
RECENT PUBLICATIONS
—ALLAN, J. A. Geology of Field Map-Area, B.C. and Alberta. [Canada
Department of Mines, Memoir 55, No. 1370, Geological Survey, Geo-
logical Series, No. 46. Ottawa, 1914.]
—American Geographical Society of New York. Memorial Volume of the
Transcontinental Excursion of 1912. [New York: Broadway at 156th
Street, 1915.]
—American Institute of Mining Engineers, Transactions of the. Vol. LI.
Containing the Papers and Discussions of the New York Meee, Febru-
ary, 1915. [New York, 1916.]
—ANREP, ALEPH v. Investigation of the Peat Bogs and Peat Industry of
Canada 1913-14. [Canada Department of Mines, ee 11, Mines
Branch, No. 351. Ottawa, 1915.]
—Batt, Max W. Petroleum Withdrawals and Restorations Affecting the
Public Domain. With 9 maps. [U.S. Geological Survey, Bulletin 623.
Washington, 1916.]
—Bancrort, J. A. Report on the Copper Deposits of the Eastern Townships
of the Province of Quebec. [Province of Quebec, Canada, Department
of Colonization, Mines, and Fisheries, Mines Branch. Quebec, r1915.]
—Bastin, E. S. Graphite in 1to15. [U.S. Geological Survey, Mineral
Resources of the United States, 1915. Part II, pp. 81-93. Washington,
1916.]
—CairneEs, D.D. Upper White River District, Yukon. [Canada Depart-
ment of Mines, Memoir so, No. 1385, Geological Survey, Geological
Series, No. 51. Ottawa,-10915.]
—CHAMBERLIN, T. C. The Origin of the Earth. The University of Chicago
Science Series. [Chicago: The University of Chicago Press, 1916.]
—Cleveland Engineering Society, Journal of the. Vol. VIII, No. 6. [Cleve-
land: 413 Chamber of Commerce Bldg., May, 1916.]
—Colorado School of Mines Quarterly. Vol. X (1915), Nos. 3, 4
—Correr, G. de P. Contents and Index of the Memoirs of the Geological
Survey of India. Vols. XXI-XXXV. 1884 to to11. [Geological
Survey of India. Calcutta, 1916.]
—Cox, G. H., anp Datce, C. L. Geological Criteria for Determining the
Structural Position of Sedimentary Beds. [School of Mines and Metal-
lurgy, University of Missouri. Bulletin, May, 1916. Rolla, 1916.]
—Cuerpo de Ingenieros de Minas del Peru. Boletin del No. 55. Informe
sobre los Trabajos de la Comision de Irrigacion de Piura por Juan N.
Portocarrero, Ingeniero Civil. [Lima: Imprenta de “El Lucerno,”
Calle de Baquijano, 767. 1907.]
498
RECENT PUBLICATIONS 499
—Daty, R. A. Problems of the Pacific Islands. [American Journal of Sci-
ence, 4th Series, Vol. XLI, No. 242. February, 1916.]
—Davis, E. F. The Registration of Earthquakes at the Berkeley Station
and at the Lick Observatory Station from April 1, 1915, to September 30,
1915. University of California Publications, Bulletin of the Seismo-
graphic Stations, No. 10, pp. 189-211. [Berkeley, March 20, 1916.]
—Davis,H.S. Spermatogenesis, Plate 8. [Museum of Comparative Zodlogy,
Harvard College, Cambridge, 1o915.]
—Dr1rr, J. S. Talc and Soapstone in 1915. [U.S. Geological Survey,
Mineral Resources of the United States, 1915. Part II, pp. 61-64.
Washington, 1916.|
—DrysDatE, C.W. Geology and Ore Deposits of Rossland, British Columbia.
[Canada Department of Mines, Memoir 77, No. 1520, Geological Survey,
Geological Series, No. 64. Ottawa, 1915.]
—ELscHNER, C. The Leeward Islands of the Hawaiian Group. Contri-
butions to the Knowledge of the Islands of Oceanica. [Reprinted from
the Sunday Advertiser, Honolulu, July 4, 1915.]
—ETHERIDGE, R. Palaeontological Contributions to the Geology of Western
Australia, Series V, No. X. [Western Australia Geological Survey,
Bulletin No. 58. Perth, 1914.]
—GeE, L. C. E. (Compiler). A Review of Mining Operations in the State
of South Australia during the Half-Year Ended June 30, 1915. [South
Australia Department of Mines, No. 22. Adelaide, 1915.|
A Review of Mining Operations in the State of South Australia
during the Half-Year Ended December 31, 1ro15. [South Australia
Department of Mines, No. 23. Adelaide, 1916.]
—Geological Magazine, No. 623, Decade VI, Vol. III, No. 5. [London:
Dulan & Co., Ltd., 37 Soho Square. May, 1016.]
—Geological Survey of Canada, Summary Report of, for the Calendar Year
1915. [Canada Department of Mines. Ottawa, 10916.]
—Geological Survey of India, Records of, Vol. XLV, Part 4, 1915. [Calcutta,
T916.|
—Glasgow University, Papers from the Geological Department, Vol. I, 1914.
[Glasgow, 1o15.|
—Go.psmitH, P. H. A Brief Bibliography of Books in English, Spanish, and
Portuguese, Relating to the Republics Commonly Called Latin-American,
with Comments. [New York: Macmillan, 1o15.]
—Grover, N. C. Surface Water Supply of the United States, 1914. Part
VIII. Western Gulf of Mexico Basins. [U.S. Geological Survey, Water-
Supply Paper 388. (Prepared in co-operation with the State of New
Mexico.) Washington, 1915.]
—Grover,.N. C., Lams, W. A., AND Hoyt, W. G. Surface Water Supply
of the United States for the Year Ending September 30, 1913. Part V.
Hudson Bay Basins and Upper Mississippi River. [U.S. Geological
500 RECENT PUBLICATIONS
Survey, Water-Supply Paper 355. (Prepared in co-operation with the
States of Minnesota and Iowa.) Washington, 1015.]
—GRroOvER, N. C., STEVENS, G. C., AND HALL, W. E. Surface Water Supply
of the United States, 1913. Part II. South Atlantic and Eastern Gulf
of Mexico Basins. [U.S. Geological Survey, Water-Supply Paper 352.
Washington, 1915.]
—HABERLE, D., uND Sotomon, W. Bericht iiber die 47. Versammlung des
Oberrheinischen geologischen Vereines zu Friedrichshafen a B. vom 14:
bis 19. April, 1914. [Jahresberichte und Mitteilungen des Oberrhein-
ischen Geologischen Vereines. N.F. 1914, Bd. IV, Heft 2, S. 69-75.
Karlsruhe: J. Langs Buchdruckerei, 1914.]
—HEnNNEN, R. V., aided by GawrHrop, R. M. West Virginia Geological
Survey County Reports, 1915. Wyoming and McDowell Counties.
With topographic and structural maps. [Morgantown, 1915.]
—HEnSsHAW, F. F., AnD DEAN, H. J. Surface Water Supply of Oregon, 1878-
1910. [U.S. Geological Survey, Water-Supply Paper 370. (Prepared
in co-operation with the State of Oregon.) Washington, 10915.]
—Hewett, D. F. Some Manganese Mines in Virginia and Maryland. [U.S.
Geological Survey, Bulletin 640-C. Washington, 1916.]
—Hrinps, H., anD GREENE, F. C. The Stratigraphy of the Pennsylvanian
Series in Missouri. With a Chapter on Invertebrate Paleontology, by
G. H. Girty. Surveyed in co-operation with the U.S. Geological Survey.
[Missouri Bureau of Geology and Mines, Vol. XIII, Second Series. Rolla,
1915.]
—Hormes, J. A., FRANKLIN, E. C., AND Goutp, R. A. Report of the Selby
Smelter Commission. [Department of the Interior, Bureau of Mines,
Bulletin 98. Washington, 1o15.]
—HvEts, F. W. The Peat Resources of Wisconsin. [Wisconsin Geological
and Natural History Survey, Bulletin XLV. Economic Series No. 20.
Madison, 1915.]
—Illinois Geological Survey. Year-Book for 1910, Administrative Report
and Various Economic and Geological Papers. Bulletin 20. Work in
co-operation with U. S. Geological Survey. [Urbana, 1915.]
Le Soldat Américain en France
By COLEMAN AND LA MESLEE
A handbook and guide for conversation for all soldiers who go
to France. Intended for those who already have some knowledge of
French but who often find themselves hard pressed for the right
word or phrase. Contains French equivalents of American military
terms for all branches of service; the necessary vocabulary of the
field hospital; words and phrases, useful in ordinary contact with
French people; and much other information to which the American
soldier in France will wish to have access.
Cloth—s54 cents postpaid
First Lessons in Spoken French for Men
in Military Service
By WILKINS, COLEMAN, AND HUSE
Prepared to meet the present emergency. A misunderstanding
of orders while in action and consequent disaster may be avoided by
a previous study of this textbook. Spoken French is presented in a
simplified manner in this convenient, pocket-size manual. Special
word-lists meet the needs of those enlisted in any branch of the
service.
Cloth—s54 cents postpaid
First Lessons in Spoken French for
Doctors and Nurses
By WILKINS, COLEMAN, AND PRESTON
Indispensable to those who desire efficiency. When health and
life are at stake, a quick grasp of the meaning of instructions is a
necessity. Special attention has been given to the preparation of
the word-lists.
Cloith—54 cents postpaid
For sale at all booksellers and news stands
THE UNIVERSITY OF CHICAGO PRESS
5825 Ellis Avenue Chicago, Illinois
FINE INKS 4x ADHESIVES
For those who KNOW
Remington Junior
Typewriter Gates
ON TEN DAYS’ FREE EXAMINATION /) brawling luke
Eternal Writing Ink
RETURN IT IF YOU WISH : ; Engrossing Ink
H ® @ 9 Taurine Mucilage
The Junior is just a boiled down | pa a | ns eeeinesdied oe
Remington, smaller, lighter, and so. simple Liquid Paste
that any one can operate it. No lessons Office Paste
needed. Vegetable Glue, Etc.
And it sells for $50. You can buy it on Are the Finest and Best Inks and Adhesives
easy terms—$10 down and $10 a month. Emancipate yourself from the use of corrosive and
iv emcking inks and adhesives andi adapt nia
ggins Inks and Adhesives. ey will bea
REMINGTON EE EWRITER COMPANY ; revelation to you, they are so sweet, clean, well
374 B aera a York : } put up, and withal so efficient.
ro way, ew OF
Send me a Remington Junior Typewriter, price $50, on free ex- 4 At Dealers Generally
amination, iat BS ener aoge thet I ney aay ee rmachines os
I oose, within ten days. ecide to purchase it, agree to a 2
pay for itin'S monthly payments of $10 each. Re CHAS. M. HIGGINS & CO., Mfrs.
Branches: Chicago, London
271 Ninth Street _ Brooklyn, N.Y.
For Practical ard
Thoroughly Efficient Work
in college and school, at public and class-
room lectures, there is no stereopticon
which meets every requirement so completely as the
Bausch’ |omb
BALOPTICON
THE PERFECT STEREOPTICON
Optical experience of more thar 60 years accounts for the superiority of the Soticall equipment of all Bausch &
Lomb models. They are, at the same time, mechanically simple and of unusual durability. The projected
images are sharp and vivid to the very edge. :
Models for lantern slides or for opaque objects (such as maps, postcards, printed pages, color charts, and
specimens) and other models combining the two, forms of projection. .A feature of great value is the recently
adapted gas-filled Mazda Lamp; it is more economical and gives far more brilliant illumination than the old-
style arc lamp.
Model C (illustrated) for lantern slides only, $35 up. Other ee a ‘$25 up.
Write today for interesting booklets and illustrated price lists. Special Terms to Educational Institutions
Bausch ¢3 lomb Optical G.
554 ST. PAUL STREET, ROCHESTER, N.Y.
NEW YORK WASHINGTON CHICAGO SAN FRANCISCO
Leading American Makers of Microscopes, Projection Apparatus, Photographic Lenses, Ophthalmic Lenses and
Instruments, Stereo- Prism Pinaeulaes and other High-Grade Optical Products.
-. eS
ee XXV
qe
NUMBER 6
JOURNAL orf GEOLOGY
SIR ARCHIBALD GEIKIE, Great Britain
A SEMI-QUARTERLY
EDITED BY
THOMAS C. CHAMBERLIN AND ROLLIN D. SALISBURY
«With the Active Collaboration of
SAMUEL W. WILLISTON, Vertebrate Balcon tology ALBERT JOHANNSEN, Petrology
STUART WELLER, Invertebrate Paleontology ROLLIN T. CHAMBERLIN, Dynamic Geology
ALBERT D. BROKAW, Economic Geology
ASSOCIATE EDITORS
JOSEPH P.IDDINGS, Washington, D.C.
Af CHARLES BARROIS, France JOHN C. BRANNER, Leland Stanford Junior University
_ ALBRECHT PENCK, Germany RICHARD A, F. PENROSE, Jr., Philadelphia, Pa.
Pm HANS REUSCH, Norway *WILLIAM B. CLARK, Johns Hopkins University
_ GERARD DEGEER, Sweden WILLIAM H. HOBBS, University of Michigan
T. W. EDGEWORTH DAVID, Australia FRANK D. ADAMS, McGill University
BAILEY WILLIS, Leland Stanford Junior University CHARLES K. LEITH, University of Wisconsin
GROVE K. GILBERT, Washington, D.C. WALLACE W. ATWOOD, Harvard University
CHARLES D. WALCOTT, Smithsonian Institution WILLIAM H, EMMONS, University of Minnesota
_ HENRY S. WILLIAMS, Cornell University ARTHUR L. DAY, Carnegie Institution
' *Deceased,
SEPTEMBER-OCTOBER 1917
STRUCTURE OF THE ANORTHOSITE BODY IN THE ADIRONDACKS H. P. Cusuine
~
ADIRONDACK INTRUSIVES - - - - - - - - - N.L. Bowen
~ADIRONDACK INTRUSIVES. - Seles - - - - - - H. P. Cusaine
A REVIEW OF THE AMORPHOUS MINERALS - - - - - Austin F. RocEers
THE CHAMPLAIN SEA IN THE LAKE ONTARIO BASIN - KirtLey F. MATHER
THE RELATIONSHIPS OF THE FOSSIL BIRD PALAEOCHENOIDES MIOCEANUS
ALEXANDER WETMORE
A STUDY OF THE FAUNAS OF THE RESIDUAL MISSISSIPPIAN OF PHELPS
COUNTY (CENTRAL OZARK REGION), MISSOURI - = JosiaAH BRIDGE
_ EVIDENCE BEARING ON A POSSIBLE NORTHEASTWARD EXTENSION OF MISSIS-
: * SIPPIAN SEAS IN ILLINOIS - - - - - - - - W. W. Davis
_ DISCUSSION OF “SOME EFFECTS OF CAPILLARITY ON OIL ACCUMULATION,” BY
A. W. McCOY - - - - - - - SG tile: - - C. W. WASHBURNE
_ PETROLOGICAL ABSTRACTS AND REVIEWS - - - = - ALBERT JOHANNSEN
REVIEWS -. ‘\- =o Ges - - oe - - - - - - Sa iit - -
ELE wUNIV Biel Y OF CHICAGO PRESS
CHICAGO, ILLINOIS, U.S.A.
THE CAMBRIDGE UNIVERSITY PRESS, Lonpon anp EDINBURGH
THE MARUZEN-KABUSHIKI-KAISHA, Toxyo, Osaxa, Kyoto, Fukuoka, SENDAI
THE MISSION BOOK COMPANY, SHANGHAL
THE JOURNAL OF GEOLOG,
EDITED BY
THOMAS C. CHAMBERLIN AND ROLLIN D. SALISBURY
With the Active Collaboration of
SAMUEL W. WILLISTON ALBERT JOHANNSEN
Vertebrate Paleontology Petrology
STUART WELLER ROLLIN T. CHAMBERLIN
Invertebrate Paleontology Dynamic Geology
ALBERT D. BROKAW
Economic Geology
The Journal of Geology is published by the University of Chicago at the University Ptess semi-
quarterly, on or about the following dates: February 1, March 15, May 1, June 15, August 1, Sept@mber 15,
November 1, December 15. {| The subscription price is $4.00 per year; the price of single topies 1s
65 cents. Orders for service of less than a half-year will be charged at the single-copy rate. {] Poktage is
prepaid by the publishers on all orders from the United States, Mexico, Cuba, Porto Kico, Panama Canal
Zone, Republic of Panama, Hawaiian Islands, Philippine Islands, Guam, Samoan Islands, Shahghai.
{| Postage is charged extra as follows: For Canada, 30 cents on annual subscriptions (total $4.36), on
single copies, 4 cents (total 69 cents); for all other countries in the Posta] Union, 53 cents on annual
subscriptions (total $4.53), on single copies, 11 cents (total 76 cents). ] Patrons are requested to make
all remittances payable to the University of Chicago Press in postal or express money orders or bank
drafts.
The following publishing houses are authorized to quote the prices indicated:
For the British Empire: The Cambridge University Press, Fetter Lane, London, E.C.4. Yearly
subscriptions, including postage, 185. 8d. each; single copies, including postage, 3s. 2d. each.
For Japan and Korea: The Maruzen-Kabushiki-Kaisha, 11 to 16 Nihonbashi Tori Sanchome,
Tokyo, Japan. Yearly subscriptions, including postage, Yen 9.00 each; single copies, including
postage, Yen 1.50 each.
For China: The Mission Book Company, 18 Peking Road, Shanghai. Yearly subscriptions,
$4.00; single copies, 65 cents, or their equivalents in Chinese money, Postage extra, if mailed direct
outside of Shanghai, on yearly subscriptions, 53 cents, on single copies, 11 cents.
Claims for missing numbers should be made within the month following the regular month of
publication. The publishers expect to supply missing numbers free only when losses have been sustained
in transit, and when the reserve stock will permit. Deliveries are not guaranteed within the war zones of —
Europe, nor in other countries where transportation is disturbed by conditions incident to war.
Business correspondence should be addressed to The University of Chicago Press, Chicago, II].
Communications for the editors and manuscripts should be addressed to the Editors of THE
JouRNAL OF GEOLOGY, the University of Chicago, Chicago, Ill.
Contributors are requested to write scientific and proper names with particular care and in citations
to follow the form shown in the pages of the Journal.
Separates, if desired, must be ordered in advance of publication. Separates of leading articles will be
printed as ordered, of which 50 (without covers) will be furnished grafzs, the actual cost of the remainder
(and covers, if desired) to be paid for by the author. The actual cost may vary, and will depend upon
the amount of work in re-making the pages into forms, presswork, paper, binding, etc. Separates
containing half-tones may be expected to cost somewhat more, the increase depending upon the number
of cuts and the amount of work required upon them.
Entered as second-class matter, March 20, 1893, at the Post-office at Chicago, IIl., under the Act of March 3, 1879
Se Ss
eS
VOLUME XXV NUMBER 6
dhes
ROM WN AL OF GEOLOGY
SBPTEMBER-OCTOLE R ror7
STRUCTURE OF THE ANORTHOSITE BODY IN .THE
ADIRONDACKS
H. P. CUSHING
Western Reserve University, Cleveland, Ohio
Introduction.—In No. 3 of the current volume of this Journal
Dr. N. L. Bowen has discussed “The Problem of the Anorthosites”’
in a very suggestive and important paper with whose general
thesis I find myself in quite hearty accord. It seems to me that
the process of formation of anorthosite, as there outlined, is quite
the most probable method yet suggested; and I am quite in agree-
ment with the explanation of the general protoclastic and granu-
lated textures which all large bodies of anorthosite exhibit.
When, however, Dr. Bowen comes to consider the universal, or
usual, field relations between anorthosite and the accompanying
bodies of syenite, and suggests in that connection a structural
relationship of these rocks in the Adirondack region, the field
facts there, as known to me, seem to be in direct conflict with
certain features of that suggestion. The chief point on which we
differ does not seem to me in any way to vitiate his main argument,
but it does seem desirable to bring it out plainly.
His argument is substantially as follows: that there is (a) an
intimate connection of syenite with anorthosite wherever the latter
is found, as shown by the abundance of rock types intermediate
501
502 H. P. CUSHING
between the two and by the rarity of exposures showing an intru-
sive relation of the one with the other; (b) an intimate association
of syenite with Grenville in many places, as contrasted with the
anorthosite areas which are comparatively free from Grenville
inclusions; also (c) the frequent occurrence of Grenville beds over-
lapping syenite in moderately undisturbed fashion, after the manner
of a roof.
Because of the latter relation Bowen argues that it is difficult
to picture the syenite and anorthosite as conventional batholiths.
He says:
It is necessary to imagine an early intrusion of a huge plug of anorthosite,
followed by an intrusion of syenite which took the form of a hollow cylinder,
circumscribing it and invading it only peripherally. All this must take place
without throwing the Grenville series into appressed folds, indeed without
significant folding of any kind. It is then necessary to imagine that erosion
removed every vestige of a roof from the small, interior anorthosite area, and
left great stretches of it throughout the broad syenite-granite belt that sur-
rounds it.
Because such relationships seem to him improbable he suggests
that the Adirondack eruptive complex consists of a sheetlike mass,
or huge laccolith, with syenite overlying anorthosite.
The purpose of this rejoinder is to point out that we are not
limited solely to the two alternatives outlined above, and that the
data obtained in the field seem to me definitely to contradict the
hypothesis that the Adirondack region is composed of a single great
sheetlike mass, with syenite overlying anorthosite and bearing all
the Grenville exposures of the region rooflike on its back.
Distribution of anorthosite and syenite-—The Adirondack
anorthosite is massed in a single great body, rudely heart-shaped,
with the apex toward the south. There are a number of small,
outlying masses, some of which have some bearing on the questions
under discussion. But in addition the continuity of the main mass
is interrupted by two considerable zulying bodies of other rock, one
in the Lake Placid region and one near Keene. Both of these are
shown on the state map. At the time when Kemp prepared this
t Jour. Geol., XXV, No. 3, 223.
THE ANORTHOSITE BODY IN THE ADIRONDACKS 503
portion of the map we were just beginning to recognize the green
syenite of the region as a later intrusive and to separate it from the
main body of the gneiss in mapping; hence comparatively little
syenite is shown as such on this map, and much of what is there
mapped as gneiss has been since shown to be syenite also. The
two inliers referred to are mapped as gneiss; but Kemp’s descrip-
tion of the rocks clearly shows that the Placid inlier consists largely
or wholly of syenite and that the same rock is represented in the
Keene inlier.‘ Both of these inliers are entirely surrounded by
anorthosite, and lie well within the mass.
By comparison with the anorthosite the syenites occur in
separate masses of smaller size, usually much smaller, and there
is a large number of such masses. Some of these directly border
the anorthosite, but there is nothing like a continuous rim of syenite
about it. These syenite masses range throughout the entire region
and, in my experience, are no more abundant near the anorthosite
than they are away from it. The anorthosite lies in the eastern
portion of the Adirondack pre-Cambrian, and its relationships to
other rocks on its eastern margin are largely hidden by down-
faulting and by the cover of Champlain Paleozoics. But to the
west and south abundant syenite ranges away to distances of 60 or
70 miles beyond the anorthosite border; and such pre-Cambrian
outliers as those at Little Falls and Middleville bear witness that it
runs an unknown distance beyond the pre-Cambrian margin under
the Paleozoic cover.
The point here made is that the syenite is spottily distributed
over the region, is no more abundant near the anorthosite than it is
away from it, and extends so far from the surface exposures of
anothorsite that the latter must be given an enormous lateral
extent underground, on the supposition that the two constitute a
sheetlike mass, with the anorthosite beneath. It is candidly ad-
mitted that this point has no particular value if Dr. W. J. Miller’s
conception of the constitution of the Adirondack pre-Cambrian
complex is the correct one. His view is that this complex con-
sists entirely of a foundation of intrusives of the anorthosite-
syenite-granite group, upon which’ fragments of the Grenville
t Bull. 21, N.Y. State Mus., pp. 55-56.
504 HOR MCU SHIN G,
cover locally persist.1 My view is that a complex of Grenville,
resting on orthogneiss, existed in the region at the time of the
intrusion of the anorthosite-syenite group, that much of this
orthogneiss still remains in the region, and that the later intrusives
broke through this complex in separate masses, instead of forming
one great body.? Obviously the presence of a great laccolith
constituting the entire region, such as Bowen postulates, is much
more possible under the former view than under the latter.
It is quite true that many of the syenite masses are much mingled
with Grenville and that the anorthosite area contrasts rather sharply
in this respect, as Bowen contends. And I am quite in accord with
his view that there has been deeper erosion in the eastern Adiron-
dacks, where the anorthosite occurs, than there has been to the
west and south, where the syenite bodies occur, and have repeat-
edly so stated. Nor am I particularly disposed to quarrel with the
view that the anorthosite mass may be laccolithic instead of batho-
lithic in structure. I do not know which it is. The mass has
certainly great thickness, since the climb up Mt. Marcy furnishes
a 3,500-foot section of pretty clean anorthosite, with no particular
indication that the entire thickness may not be vastly greater;
then allowance must be made for at least an equal thickness of
gabbro and pyroxnite underneath and for an unknown thickness
of overlying syenite, since eroded away. Nevertheless, a sheet
structure is entirely possible.
Differentiation of the anorthosite body.—If I have correctly under-
stood Dr. Bowen’s interpretation of the structure of the region—a
sheetlike igneous mass, composed of probable gabbro below, then
anorthosite, and finally a cover of syenite and granite, the cover
full of fragments from the Grenville roof, and with a certain
amount of disturbance occurring during the freezing of the mass,
whereby liquid syenite is brought into lateral contact with solid
anorthosite—his argument seems to me to imply, or to require, that
this sheet was at least equal in size to the present pre-Cambrian
area of northern New York and that anorthosite must everywhere
underlie syenite. The field evidence, however, seems to me to
t Bull. Geol. Soc. Am., XXV, 243-64. ;
?Am. Jour. Sci., XX XIX, 288-94.
THE ANORTHOSITE BODY IN THE ADIRONDACKS 505
demonstrate that the full girth of the anorthosite intrusion is
represented by the dimensions shown on the present maps and that
the outlying syenite bodies represent distinct and slightly later
intrusions.
Dr. Bowen’ discusses the ‘Intimate Relation of Syenite and
Anorthosite’’? and makes the following statement:
This aspect of the anorthosite, i.e., its intimate connection with the syenite,
is emphasized in the area as a whole, where, in spite of fairly good exposures,
only one other locality showing the intrusive relation of syenite to anorthosite
has been found, but where, on the other hand, types intermediate between the
two are rather commonly found.
It is chiefly these two points of his paper which I wish to discuss,
since my field experience is quite antagonistic to them. I have
myself published several localities where dikes of syenite cut the
anorthosite, and on the next page of his paper Bowen quotes W. J.
Miller as authority for the statement that dikes of syenite cutting
anorthosite occur in the Placid region. My thesis is that the
general differentiation zm sztw shown by the anorthosite is into
anorthosite-gabbro and gabbro, and not into syenite; that such
intermediate rocks as do occur are chiefly intermediate between
syenite and gabbro, instead of between syenite and anorthosite;
and that the demonstrable source of these latter intermediate rocks
in many cases, if not in most or all of them, is assimilative attack
of a later intrusion upon an earlier, and is not differentiation
im situ.
The boundary of the anorthosite is in part along faults. Where
unfaulted the anorthosite is always found to grade into anorthosite-
gabbro, and this into gabbro as the boundary is approached. This
change is depicted upon the Long Lake and the Elizabethtown
quadrangle maps and occurs also in all other parts of the region in
which I have any acquaintance with the boundary. Daly has
interpreted this as a chilled border of the anorthosite, and in my
judgment this is not only the most reasonable, but in fact the only
satisfactory, explanation that can be made of it.? If this be true,
VOPACie pr 2ut.
2 Igneous Rocks and Their Origin, p. 240.
506 H. P. CUSHING
it follows of necessity that the anorthosite is a differentiate im situ —
from a gabbro intrusion and that the chilled border determines for us
the original size of the mass at the depth represented by the present
erosion surface—in other words, that this particular anorthosite
mass cannot be regarded as spreading out underneath the outlying
syenite masses and extending throughout the region. If it be
argued that the anorthosite body, while cooling, developed a syenite
cover, since removed by erosion, I would state that I think this
very probable, and would even go so far as to suggest that the Placid
and Keene inliers of syenite may be remnants of this cover. If so,
it would be in their vicinity that true transitional rocks between
syenite and anorthosite would be most likely to occur. Whether
they so originated, or represent plugs of syenite rising through the
~ anorthosite, can be determined only in the field with aid of favorable
exposures, if any such exist. But even this gives no aid in explana-
tion of the outlying syenite masses."
The syenite-anorthosite boundary across the Long Lake
quadrangle suggests intrusive attack of syenite upon anorthosite
for its entire length. Along it the syenite develops a basic border
of its own, which I have elsewhere endeavored to show is due to
assimilation of gabbro and anorthosite gabbro by the molten
syenite. In places the syenite thrusts deep salients into the
anorthosite, and an excellent sample may be seen on the Long
Lake map, coming down to the Raquette River just north of
Raquette Falls. It cuts into the anorthosite body to a depth of
two miles, cutting out much of the gabbro and anorthosite-gabbro
border, though these appear in full width on both sides of the salient.
Within it are several inclusions of anorthosite gabbro, five of which
are of sufficient size to be delineated on the map. Each inclusion
has an aureole of very basic syenite, grading away imperceptibly
into the normal rock. These are remnants of the anorthosite-
gabbro border which was there before the syenite salient was thrust
in, and which has escaped the utter digestion experienced by the
remainder.
tIt should be noted that the transitional anapersites described by Bowen are
from the Placid region (op. cit., pp. 221-22).
2 Bull. Geol. Soc. Am., XVIII, 477-92; Bull. 115, N.Y. State Mus., pp. 478-82.
THE ANORTHOSITE BODY IN THE ADIRONDACKS 507
In addition the anorthosite is cut by dikes of syenite in several
localities, some of them four or five miles in from the anorthosite
border. The field evidence seems clear that the anorthosite had
solidified, with a chilled border, and had then been attacked from
the side by a mass of molten syenite, which in places cut deeply
into it. Along the contact a basic border phase of the syenite was
produced, which is not a chilled border because found only along that
part of the syenite boundary which 1s in contact with the anorthosite,
and hence must be due to the assimilative incorporation of anortho-
sitic material. The product is an intermediate rock, but inter-
mediate between syenite and gabbro rather than between syenite
and anorthosite. It differs from the normal syenite chiefly in its
large content of ferromagnesian minerals rather than by pro-
nounced difference in the character of the feldspar. It somewhat
resembles the gabbro, but ability to distinguish the two is quickly
attained in the field.
Dr. Bowen’s suggested interpretation of these relations is that
disturbance occurred during solidification of the sheet-like mass,
after the anorthosite had become practically solid, but while the
overlying syenite was still fluid, faulting the one against the other
and thus permitting the fluid to laterally attack the solid, giving
rise to the intrusive features found in the field. This is a possible
cause of such relationships, but it seems to me that the presence
of the chilled gabbro border of the anorthosite is fatal to its applica-
tion in this particular case. That border seems to me to indicate
that this is the original size of the anorthosite mass; that it cannot
therefore extend westward underneath the bordering syenite; that
it cannot possibly underlie the great number of other syenite bodies
which range away for distances exceeding 50 miles to the west and
south.
The presence of anorthosite outliers in the region somewhat
complicates the problem, and might be thought to lend support.
to Bowen’s conception of the structure. The largest of these known
to me is that at Rand Hill, Clinton County, which I described years
ago. This lies 20 miles distant from the nearest part of the
main body, at Keeseville, and seems to me to represent a distinct
intrusion, though in all probability an offshoot from the same
508 » _. HP. CUSHING
parent-mass below ground. The other known outliers are all small,
are all composed of anorthosite gabbro, lie within a distance of to
miles from the anorthosite boundary, and are either demonstrable
or probable inclusions in the syenite or else are dikelike or pluglike
offshoots from the main mass. So far as I know the evidence, they
do not at all require belief in the greater extent of the anorthosite
mass underground.
Concluston.—While, therefore, I am quite in accord with
Dr. Bowen in the belief that the gabbro, anorthosite, syenite, and
(in part) granite bodies of the Adirondacks are all differentiates
from a common parent-magma and are closely akin in age, I do
not believe that the present surface exposures can be successfully
explained as constituting one great igneous body. The anorthosite
mass arose to its present position as a gabbro magma, developed a
chilled border, differentiated with production of anorthosite and
quite possibly overlying syenite, and solidified. The overlying
syenite has since been eroded away, except for the possibility that
the Placid and Keene inliers may represent portions of it. In so
far I can follow Bowen without trouble. But to account for the
outlying syenite and granite bodies away from the anorthosite I
think that we must resort to the conception of at least one other
body of magma, probably of several others, which went through a
similar differentiation well below the present surface and from whose
upper parts bodies of molten syenite were pushed upward. Some
of these came up along the margins of the solidified anorthosite
mass while it was still hot and produced the contact relations which
we find today.
It is not safe to say, at the present time, that the floor of the
entire Adirondack region is constituted of representatives of this
one igneous group. It is quite true that the Grenville remnants
in the region always rest on igneous rocks which bear an intrusive
relation to them. But in my view there are considerable masses
of orthogneiss present much older than the rocks of the anorthosite-
syenite group; in places the Grenville rests on these, and it is not
at all certain that they rest on the younger intrusives; and in many
parts of the region these rocks border the anorthosite, as, for
example, along Cold River on the Long Lake quadrangle. The
THE ANORTHOSITE BODY IN. THE ADIRONDACKS 509
conception of a cylinder of syenite enfolding anorthosite is therefore
neither a necessary nor a true one; rather, there are a number of
separate syenite masses.
ADIRONDACK INTRUSIVES
N. L. BOWEN
Geophysical Laboratory, Carnegie Institution of Washington
In his paper on the “Structure of the Anorthosite Body in the
Adirondacks” Professor Cushing offers some objections to the
interpretation of Adirondack igneous geology that was given by
me in the paper “The Problem of the Anorthosites,”’ and he has
kindly asked me to comment upon his objections. It naturally
gives me considerable satisfaction that an investigator with Pro-
fessor Cushing’s broad experience of Adirondack geology should
accept the more important and vital aspects of my interpretation
of the genesis of Adirondack igneous types. I therefore find myself
disinclined to object very vigorously to his remarks on features
of Adirondack structure concerning which he finds it necessary
to disagree with me. This is especially true since it would be
presumptuous on my part to differ from him on any point involving
actual knowledge of field facts. Nevertheless, there seem to be
certain questions of interpretation on which there is room for
alternative views.
The common, basic border phase of the anorthosite Professor
Cushing considers fatal to the idea of the extension of that rock
type laterally as a sheetlike mass beyond the limits of its present
exposure. He accepts Daly’s interpretation of this border phase
as a chilled portion and considers that this phase must ‘be the
outer limit of the anorthosite. I, too, accept Daly’s interpretation
of the basic border, but consider that it is not necessarily an outer
limit; it may be an upper limit, or rather a former upper limit.
It may therefore represent a chilled upper portion of a laccolithic
mass extending far beyond the limits of its present exposure.
It is perhaps necessary to go into this matter in greater detail,
and, in order that this may be done, mention will first be made of
510 N. L. BOWEN
a much simpler example of the same phenomenon. In the Palisade
diabase of New Jersey gravitative differentiation has taken place
with the result that there has been formed in the lower layers an
olivine-rich diabase and in the upper layers more acidic types, in
local patches verging upon granite. At the upper border, however,
a more basic phase occurs which contains a small amount of olivine
and represents the original magma quickly chilled and undiffer-
entiated. In this body of moderate dimensions all the differentiates
have remained in position, except that the acidic phase may be
injected occasionally into the more basic varieties as aplitic dikes.
When this occurs, the acidic phase has been noted to exert a par-
ticularly strong corrosive or recrystallizing action on the basic
phase.
While the differentiates are of other types in the case of the
Adirondack complex, I believe that in a broad way the relations
are substantially the same, the principal complicating circumstance
being the prominence of reintrusion of the later liquid, the syenite.
The gabbro border phase I believe, with Daly and Cushing, to be
a chilled border, and, while this matter was not discussed in con-
nection with the Adirondacks, mention was made of such chilled
phases on page 213. In an undisturbed mass the syenite would
everywhere lie immediately below this basic phase if the mass had
also a very regular contact. However—and this brings us to
another of Professor Cushing’s objections—if the mass had an
irregular upper contact, the syenite need be present only in the
re-entrants of the roof and need not therefore form a continuous
border about the anorthosite. Add to this the fact that the syenite
has been disturbed and re-intrusion has occurred, and I think that
this fact will become still more obvious. It must be confessed that
Professor Cushing was perhaps justified in considering a continuous
syenite body a necessary consequence of my hypothesis on account
of the diagrams that were offered in illustration of the conception.
But these were intended to represent in a diagrammatic way the
conditions under which the various types were generated, and not
to give a picture, except in a generalized way, of the actual distribu-
tion of types in the Adirondacks at present. It is recognized that
reintrusion of the syenite occurred, resulting in satellitic bodies
THE ANORTHOSITE BODY IN THE ADIRONDACKS 511
at higher horizons in the Grenville, though much of it remained
substantially where generated—enough, perhaps, to justify the
statement that the Adirondack complex is ‘essentially a sheetlike
mass with syenite overlying anorthosite.’”’? Whether distant syenite
masses are to be regarded as related to the anorthosite I cannot say.
As a consequence of reintrusion, invasion of the anorthosite
by syenite, in so far as this occurs, is especially likely to be true of
the basic border phase, the anorthosite-gabbro or gabbro, and, after
the manner of the acidic phases in the Palisade intrusive, it may be
expected that the syenite will exert a strong corrosive or resorbing
action on these basic differentiates, such as that described by
Professor Cushing from Long Lake. While, therefore, the syenite
would be pushed up from below into and beyond the basic phase
of the anorthosite, it is considered that the syenite came into being
at a higher level than the anorthosite proper. This is not incon-
sistent with the occasional occurrence of dikes of syenite in the
anorthosite, for a splitting of the solidified anorthosite would permit
the formation of such dikes from an overlying liquid syenite as
readily as from a deeper-seated mass. Professor Cushing is able
to bring into court more examples of these dikes than I had supposed
were known, but I think that it must be admitted that in much
of the quadrangle work the syenite is considered later than the
anorthosite solely on the basis of his findings in the Long Lake
quadrangle. This might be considered as due to failure of exposure,
but in the same areas there is no lack of evidence of the invasion
of the Grenville by syenite. I consider it likely, therefore, that
the syenite does not invade the anorthosite in exactly the same way,
but is largely transitional into it, although, being of somewhat later
consolidation, it may send dikes into the anorthosite on occasion.
My observations are admittedly limited, but I do not think that
the intermediate types to be seen at Lake Placid are formed by
interaction of the two types, an action of which Professor Cushing
finds abundant evidence at Long Lake. The Placid types are quite
definitely intermediate between syenite and anorthosite, not
between syenite and gabbro, as are Cushing’s reaction types.
In conclusion, I would state that, while Professor Cushing has
raised legitimate objections and there is certainly room for difference
512 H. P. CUSHING
of opinion, it still seems to me to be advisable to keep an open
mind on the possibility that the syenite and anorthosite occur
“substantially as layers with the syenite above.” It is especially
desirable in view of the fact, recognized by Professor Cushing, that
the anorthosite occurs in the more deeply eroded portions and the
syenite principally at higher horizons, an arrangement not easily
reconciled with the opinion that the syenite pushes up into the
anorthosite from below.
ADIRONDACK INTRUSIVES
H. P. CUSHING
I am greatly indebted to Dr. Bowen for his additional contribu-
tion to this discussion. I take the liberty of considering briefly
the points he brings out.
He suggests that the chilled gabbro border of the anorthosite
is not a lateral border but a remnant ofian upper one. It is very
difficult for me clearly to visualize the structure of the region on this
view. It is, roughly, about 1oo miles across the mid-Adirondack
region from east to west, and, again roughly, the easterly half of
this distance is occupied by pretty clean anorthosite, and the
westerly half contains a great number of syenite bodies and no
anorthosite at all. The chilled gabbro border is about midway
of the region. If it is a chilled upper portion of a laccolith, consist-
ing of pyroxenite and gabbro below, then anorthosite, then syenite,
and, finally, the chilled gabbro roof, since tilted so that the present
erosion surface cuts it at a considerable angle, it is necessary to
conceive that this chilled upper surface passes below ground in
the westerly direction and into the air to the east. Under such a
view the present-day syenite masses of the west must have broken
through this cover to reach their present position, and there is no
particular difficulty in imagining that they did so. But under
this view it seems to me necessary that we should also find syenite
to the east of the chilled border and close to it—that syenite which
formed as a differentiate in the upper part of the chamber, under-
neath the chilled upper surface. Even if the upper part was very
THE ANORTHOSITE BODY IN THE ADIRONDACKS 513
irregular, as Bowen suggests, so that the syenite differentiate was
in separate masses instead of in a continuous sheet, there should
still remain considerable masses of syenite within the chilled border
and underlying it, if this border was an upper instead of a lateral
one. We should find gabbro passing into syenite and this into
anorthosite. I know of no such syenite masses within the gabbro
border anywhere in the region; and to explain their absence we
should be, it seems to me, forced to the conclusion that, owing to
disturbance during a late stage of consolidation, every particle
of this syenite differentiate in the upper part of the body was forced
out through the roof to higher levels, letting the chilled gabbro
down upon the anorthosite. This is perhaps possible, but certainly
very unlikely, and, moreover, it would leave unexplained the usual
slow and even gradation from border gabbro into anorthosite
gabbro and of this into anorthosite as we recede from the border—
a feature which seems to me convincingly to suggest a lateral border
rather than an upper one.
In the Palisade sheet, utilized as an illustration by Bowen, the
acidic types lie directly underneath the upper chilled border. In
the Adirondacks they lie without and above rather than within
and below this border, as they should do on the sheet conception.
None of the outlying masses of syenite known to me show any
sign of a chilled border of gabbro as if they were upper parts of a
single large body. They all seem rather of the type of injections
upward from some large mass of magma below. It is to be under-
stood that I am not objecting to the laccolithic conception, but
to the conception of a single laccolith occupying the entire region.
If we regard the anorthosite mass as a single mass, its margins
shown by the chilled border, attacked shortly after its formation
by masses of syenite magma, which arose from one or more separate
and deeper bodies to the west and which came up along the margin
of the anorthosite, not through it, we obtain an explanation of the
abrupt transition from anorthosite to syenite territory which ob-
tains in the region and we are free from the difficulties which have
just been discussed.
Dr. Bowen’s contention that the present-day ideas in regard
to the time relationship between syenite and anorthosite are chiefly
514 H. P. CUSHING
due to my observations on the Long Lake quadrangle is correct,
but needs some comment in order not to be misleading. The
anorthosite district is rugged, wooded, and difficult. Very little
of it has been mapped in detail on quadrangle maps. Moreover,
the marginal types are weaker and less likely to be well exposed.
It is as yet unsafe to argue that the Long Lake phenomena are
exceptional because they have not been shown elsewhere.
I regret that Dr. Bowen did not discuss my suggested explana-
tion of the Placid syenite inlier as a remnant of the overlying
syenite differentiate of the anorthosite. It is there that he found
his rocks transitional between syenite and anorthosite, and it is
there that they would be expected, if the inlier is such a remnant.
These rocks are not on the border, but are within the mass. If a
chilled border is lacking between them and the adjacent anorthosite,
then the occurrence would seem most easily explained on this
theory.
In conclusion, then, the relations shown in the field along the
western border of the anorthosite seem to me to indicate that the
anorthosite is one body and the syenite masses to the west and south
belong to one or more separate and slightly later bodies. While
I welcome Dr. Bowen’s theory of the formation of the anor-
thosite and syenite, I can neither agree with, nor see the neces-
sity of, his idea that all these intrusives of the region are parts of
one single laccolithic body. As I picture it, they must represent
separate upwellings from one or more deeper-seated magmas.
The western border of the anorthosite marks the line of division
of the region into contrasted halves—anorthosite to the east,
separate syenite masses to the west; and the particular syenite
masses which happened to adjoin the anorthosite differ in no
particular from those more remote, except in the one that they
have assimilated some anorthosite at the contact of the two rocks.
Nothing is to be gained by assuming that the adjoining masses
belong with the anorthosite body and by attempting to make a
separation between them and those more remote. Such a separa-
tion would be purely arbitrary, whereas the other divisional line
is sharp and obvious.
A REVIEW OF THE AMORPHOUS MINERALS
- AUSTIN F. ROGERS
Stanford University, California
Mineralogists generally have neglected the study of naturally
occurring amorphous substances. None of the modern miner-
alogical textbooks or treatises give an adequate treatment of the
amorphous state or condition. Crystals are treated at great length,
but the amorphous state is usually given only a paragraph or two."
The reasons for this neglect on the part of mineralogists are
apparent. Crystals, with their great variety of form and physical
properties, offer a more attractive field for study. Crystalline
material so greatly predominates over amorphous material in the
earth’s outer shell that it is regarded as typical of the solid state.
In selecting material for chemical analysis crystals are selected as
far as possible, for crystallization is Nature’s great method of pro-
ducing pure inorganic substances.
While the mineralogist is primarily a crystallographer and while
crystallography is left largely in the hands of the mineralogist, we
need to be reminded that mineralogy and crystallography are by
no means synonymous. Crystallography deals with crystals pro-
duced in the laboratory as well as with mineral crystals. Min-
eralogy deals with all homogeneous, naturally occurring, inorganic
substances, whether crystalline or amorphous. The two fields
overlap but do not coincide.
In view of the recent advances in colloid chemistry, the mineralo-
gist can no longer be excused for his neglect of the study of the
amorphous state. Although the science of colloid chemistry has
been developed largely by chemists, Breithaupt, one of the early
mineralogists and perhaps the greatest of the old natural-history
t Knop, however, in his System der Anorganographie (Leipzig, 1876), and Doelter,
in his Physikalisch-chemische Mineralogie ((Leipzig, 1905), both treat the amorphous
state at some length.
515
516 AUSTIN F. ROGERS
school of mineralogists, recognized colloids under the name poro-
dine substances as early as 1817.1 He called attention to the
fundamental differences between the crystalline and amorphous
conditions and divided amorphous substances into two groups: the
hyaline (glasses) and porodine (“guhren” or gels). Breithaupt
knew of the researches of Graham before the publication of
the latter’s discovery of dialysis and stated that his group of
‘“ouhren”’ was identical with the laboratory products of Graham.
The application of colloid chemistry to mineralogy has been
pointed out by Cornu and others in a series of papers published in
1909”, and more recently Marc and Himmelbauer’ have contributed
an excellent summary and bibliography of the whole subject.
CONFUSION OF THE TERMS AMORPHOUS AND COLLOIDAL
While the study of the amorphous condition involves a study
of the colloidal or ‘‘dispersed”’ state, the terms amorphous and
colloidal are not synonymous. Amorphous is the broader term.
Ordinary glasses are amorphous but not colloidal, although some
varieties of glass may be colloidal (e.g., opalescent glass). Col-
loids are microheterogeneous systems made up of two phases, one |
dispersed through another.
Most of the amorphous minerals are hardened hydrogels. The
water present is usually considered to be adsorbed water. It
may be adsorbed water when the colloid is first formed (even this
fact is doubted by Robertson),‘ but after hardening, the water may
be present in solid solution.’ Hyalite opal, for example, is appar-
ently a microhomogeneous substance, and there is no reason why
it may not be looked upon as a true solution of water in silica.
Hence it is possible, and even probable, that the mineral hydrogels
are not, properly speaking, colloids, but only colloidal in origin.
«Cornu, Zeit. f. Chem. u. Ind. d. Kolloide, IV (1909), 300-4; Hunt, Systematic
Mineralogy, New York, 1891, p. 10.
2 Tbid., IV, 15, 89, 187, 188, 189, 275, 285, 291, 295, 298, 300, 304, 306.
3 Fortschritte der Min., Krist., u. Petrog., III (1913), 11, 32.
4 Zeit. f. Chem. u. Ind. d. Kolloide, II (1908), 40.
5 The fluorin content of collophone mentioned later in this paper is an argument
in favor of the solid solution theory of hydrogels.
A REVIEW OF THE AMORPHOUS MINERALS 517
Wolfgang Ostwald says: “‘The colloid solutions .... should
always be meant when colloids in general are under discussion,’’!
and also ‘‘ when we speak of a ‘colloid’ we nearly always mean one in
this condition, in other words, one in the sol condition.’”
Biitschli described the gels as possessing a honeycomb struc-
ture, but according to Bachmann, this is simply an optical effect.
By means of the ultramicroscope the latter proved that the struc-
ture of the solid gel of silica is extraordinarily fine and for the most
part amicroscopic (i.e., made up of amicrons).
Whether the hardened gels are colloidal or not, we are safe
in calling them amorphous.
There is a tendency on the part of some modern mineralogists
to use the term colloid, not only for the hydrogel minerals, but
also for microcrystalline substances of colloidal origin. The recog-
nition of colloidal structures in the study of minerals, rocks, and
ores‘ is important, but the fundamental differences between
amorphous and crystalline minerals should not be lost sight of. In
the identification of minerals the mineralogist is concerned with
amorphous and crystalline substances, and not primarily with
colloidal structures. The confusion of the terms colloidal and
amorphous is apparent in a number of recent mineralogical papers.
For example, CornuS classifies chrysocolla as the ‘‘gelform”’ of
dioptase, when, as a matter of fact, chrysocolla is crystalline. Its
amorphous equivalent is another mineral recognized in this paper
for the first time. The use of the term metacolloid proposed by
Wherry? for microcrystalline substances of colloidal origin will do
much to clear up the difficulty. Cornu’s term “‘gelform”’ is ambigu-
ous, and as an illustration let us take the silica minerals. Opal is
listed as the “‘gelform”’ of chalcedony by Cornu and chalcedony as
a “‘krystalloidform.” But both of the minerals have colloidal
* Handbook of Colloid Chemistry (Eng. trans. of 3d edition by Fischer, 1915), p .24.
2 Ibid., p. 40.
3 Zeit. f. anorg. Chemie, LX XIII (1911), 125.
4 For a discussion of colloidal structures in ores see Krusch, Zezt. f. prakt. Geol.,
21. Jahrgang (1913), 506-13.
5 Zeit. f. Chem. u. Ind. d. Kolloide, IV (1909), 17.
6 Jour. Wash. Acad. Sci., IV (1914), 112.
518 AUSTIN F. ROGERS
characteristics, and if we treat them as Cornu treats chrysocolla
they would both be considered as colloidal or “gelforms”’ of quartz.
In the case of silica minerals, for example, ambiguity is avoided if
we call opal amorphous silica and chalcedony metacolloidal silica."
CRITERIA FOR THE RECOGNITION OF AMORPHOUS SUBSTANCES
To the mineralogist is committed the task of describing and
defining all of the definite, homogeneous, naturally occurring sub-
stances, whether they be crystalline or amorphous.
How may amorphous substances be distinguished from crystal- .
line substances? On the face of it this seems to be easy, but, as a
matter of fact, the problem is often very difficult. Crystalline
substances may be defined as those having discontinuous vectorial
properties? and amorphous substances as those not having such
properties, but the actual determination of whether a given sub-.
stance has discontinuous vectorial properties or not may be very
difficult.
From the standpoint of physical chemistry amorphous solids
are liquids. Now, the shape of a liquid unaffected by gravity or
other external influence is spherical, and so we often find the
hydrogel minerals in spherical, botryoidal, reniform, stalactitic,
and mammillary forms. These forms intergrade, so that one is
often at a loss to know which term to use. I therefore propose the
term colloform for the rounded, more or less spherical, forms assumed
by colloidal and metacolloidal substances in open spaces. Some
crystalline, not merely microcrystalline, minerals, such as smithson-
ite, also occur in colloform crusts, and it should be emphasized
that this term refers only to the shape or form, and not to the con-
dition of the material.
Colloform minerals may be either amorphous or crystalline,
while, on the other hand, minerals occurring in euhedral crystals
may be amorphous alteration products of original crystals, for
example, malacon, which is a pseudomorph after zircon. Yttrotan-
talite, thorite, allanite, gadolinite, homilite, and yttrocrasite all
occur in euhedral tetragonal, orthorhombic, or monoclinic crystals.
* Chalcedony may be either a distinct mineral or a variety of quartz.
2 Friedel, Legons de Cristallographie (Paris, 1911), p. 2.
A REVIEW OF THE AMORPHOUS MINERALS 519
Yet in many cases the material is optically isotropic and we appar-
ently have the amorphous equivalents of these minerals occurring
as pseudomorphs after the original crystalline minerals. The typi-
cal structures of crystalline aggregates, such as fibrous, lamellar, etc.,
do not necessarily indicate crystallinity, for these structures may
be remnants of an original crystalline condition, now consisting of
amorphous material. Pyrolusite, for example, is probably an
amorphous manganese dioxid produced by the dehydration of
crystalline manganite, which accounts for its fibrous structure.
In the absence of cleavage and other direct proofs of crystal-
linity we must rely largely upon optical tests for transparent and
translucent minerals. A serious difficulty confronts us here, for
isometric crystals, as well as amorphous solids, are optically iso-
tropic, and anhedral isometric crystals without cleavage may be
confused with amorphous substances of similar appearance.
Still another difficulty lies in the fact that many amorphous
substances are doubly refracting. This is especially true of collo-
form crusts, and the double refraction here is due to strains set
up in the hardening of the gel. The hyalite variety of opal prac-
tically always shows double refraction. The birefringence of an
amorphous mineral is usually very weak, but in some cases it reaches
an appreciable amount. In a specimen of phosphorite from Lassa
Island containing both dahllite and collophane the double refrac-
tion of colloform bands of amorphous collophane is greater than that
of the corresponding crystalline dahllite. Double refraction due
to strain can usually be distinguished from the double refraction of
optically anisotropic crystalline substances by its lack of uniformity.
For opaque minerals etching experiments are perhaps the most
satisfactory tests to try in the absence of evident crystalline struc-
ture. Tolman* has recently described metacolloidal chalcocite.
The determination was made by examining polished surfaces
etched by nitric acid with the metallographic microscope.
From the foregoing discussion it is evident that in many cases
the scalar properties must be used to distinguish amorphous and
crystalline substances, and it should be emphasized that the optical
properties are also scalar for isometric crystals.
Bull. 110 Am. Inst. Min. Eng., 1916, p. 410.
520 AUSTIN F. ROGERS
Among the scalar properties are specific gravity, specific heat,
fusibility, solubility, and also the index of refraction for both iso-
metric crystals and amorphous substances. Now, any one of these
properties is somewhat different™ for an amorphous substance and
the corresponding crystalline substance, i.e., one with the same, or
approximately the same, chemical composition. The solubility
and fusibility are not easy to determine accurately, and this is
often true of the specific gravity.
The determination of the index of refraction, and not the
presence or absence of double refraction, furnishes the most gener-
ally available means of identifying a given amorphous mineral.
Irregular grains of garnet in the form of sand, for example. are
identified as garnet, not because it proves to be crystalline, but
because of the isotropic character, high index of refraction, absence
of cleavage, pink color, etc. We know it to be crystalline simply
because it is garnet. An amorphous mineral corresponding to
garnet would have a lower index of refraction. It might be difficult
to prove that such a mineral is amorpnous in the first place, but if
this fact were once established the mineral could be distinguished
from garnet or, more accurately speaking, from one of the members
of the garnet group by its index of refraction. The index of re-
fraction, however, is sometimes misleading. For example, lussatete
is a fibrous variety of silica probably identical with chalcedony, yet
it has the index of refraction of opal. The explanation is that
minute fibrous aggregates of chalcedony have gradually crystallized
out of an amorphous mass of opal.
Many of the amorphous minerals may be distinguished from
their crystalline equivalents by the presence of water, which seems
to be almost universally present in the amorphous minerals.
THE GENERALLY RECOGNIZED AMORPHOUS MINERALS
Comparatively few amorphous minerals are recognized in
standard works on mineralogy. In Dana’s System of Mineralogy
(6th edition, 1892) with its three appendixes (1904, 1909, 1915), for
example, approximately one thousand minerals are given the rank
tSee Knop, op. cit., p. 8.
A REVIEW OF THE AMORPHOUS MINERALS 521
of distinct species, and of these only seventeen" are listed as
amorphous. Some of these are probably synonyms and some
are undoubtedly crystalline. Of the amorphous minerals
described since the appearance of Dana’s System none has
been given the rank of mineral species in the appendixes of that
work.
The principal reason why so few amorphous minerals are listed
is because the amorphous equivalents of crystalline minerals, with
the single exception of opal, are not recognized as distinct mineral
species. We find, for example, crystalline cupric oxid (tenorite)
united with amorphous cupric oxid (melaconite). Amorphous
limonite is united with its fibrous and crystalline equivalent.
Crystalline and amorphous ferric oxid are both included under the
name hematite.
ARE AMORPHOUS MINERALS TO BE RECOGNIZED ?
The chemical composition of most amorphous minerals is not
as definite as that of the average crystalline mineral, but since the
discovery of the variability in the composition due to solid solution
of such well-crystallized minerals as pyrrhotite and nephelite, we
can no longer insist that the term “mineral species”? be confined
to those of definite chemical composition. Many of the amorphous
minerals approach closely in chemical composition the correspond-
ing crystalline mineral, as was first emphasized by Cornu,? who
called these minerals “‘pseudo-stéchiolithe.” Practically all the
amorphous minerals contain water, even if the corresponding
crystalline minerals are anhydrous, but this water or the excess
water over that present as hydrion or hydroxyl is probably not
essential. The myeline from Rochlitz, Saxony, for example, is
probably the amorphous equivalent of kaolinite, yet it contains
practically the same amount of water.
Some of the amorphous minerals are definite enough to be
recognized, though we must, of course, allow more latitude in
1 These are opal, collophanite, bindheimite, szmikite, deweylite, genthite, garnier-
ite, spadaite, saponite, glauconite, cimolite, montmorillonite, allophane, collyrite,
schrétterite, chloropal, and hisingerite.
2 Zeit. f. Chem. u. Ind. d. Kolloide, IV (1909), 15, 89.
522 AUSTIN F. ROGERS
chemical composition and physical properties than with crystalline
minerals. The physical properties of many of the amorphous
minerals can be determined as completely as those of many of the
massive non-cleavable isometric minerals, for in the absence of
cleavage and crystal form all the available properties for determina-
tion are scalar.
Most authorities admit an amorphous mineral to the full rank
of species if it has no crystalline equivalent, but discard those with
crystalline equivalents. This, I believe, is inconsistent, and it
would seem more logical to refuse admittance to an amorphous
mineral until its crystalline modification is described. It is also
inconsistent to recognize opal and not the other amorphous equiva-
lent of crystallized minerals. Opal is not a definite hydrate of silica,
but is silica with dissolved or adsorbed water. Other amorphous
minerals also contain dissolved or adsorbed water and bear the
same relation to crystalline equivalents that opal does to quartz.
As shown by von Weimarn,' the colloidal condition is a general
property of matter. No one can doubt that the amorphous and
crystalline conditions are fundamentally different. Any given
substance possesses a different energy content in these two condi-
tions.
If it is admitted that the properties of amorphous minerals are
sufficiently distinct, and we admit this when we use such names as
opal, pitticite, allophane, etc., we must assign the known amorphous
equivalents of crystallized minerals the rank of independent mineral
species. The first step in this direction was taken, I believe, by
Cornu? in t909. He introduces the names kliachite, stilpnosider-
ite, gelvariscite, gelfischerite, geldiadochite, gelpyrophyllite, etc.,
as names of the amorphous equivalents of hydrargillite, limonite,
variscite, fischerite, diadochite, and pyrophyllite, respectively.
This, in my opinion, is one of the important advances in systematic
mineralogy. Although sound from the standpoint of physical
chemistry, this principle has not been generally adopted. We are
conservative, and even desirable changes are slow in adoption, but
it seems strange that such names as cliachite or kliachite are not
* Zur Lehre von den Ziistanden der Materie (Leipzig, 1914).
2 Zeit. f. Chem. u. Ind. d. Kolloide, TV (1909), 15.
A REVIEW OF THE AMORPHOUS MINERALS 523
adopted when so many new names of mere varieties or mixtures
of older minerals’ are apparently welcomed.
NAMES FOR AMORPHOUS MINERALS
If it be granted that amorphous minerals deserve recognition,
then names of some kind are necessary. Are these to be distinctive
names or modifications of the names of the corresponding crystal-
line minerals? Cornu used the prefix ‘‘gel’”’ with the crystalline
modification (e.g., gelvariscite, for the amorphous equivalent of
variscite). Tucan? employed a similar device, except that “gel”
came after the root name instead of before it (e.g., hematogelite
for colloidal ferric oxid). Wherry? proposed that the Greek letter
x (the abbreviation of xo\da) be used as a prefix to the crystalline
compound (e.g., x-limonite for stilpnosiderite).
A serious objection to all these proposals lies in the fact that
the amorphous mineral is not related to one polymorphous modi-
fication any more than to another. Most mineral substances are
known in but one crystalline modification, but other modifications
may be found in the future, as polymorphism seems to be a general
phenomenon of nature.
The name of any crystalline mineral connotes certain crystal
forms and physical properties as well as a given chemical composi-
tion. It is absurd, then, to speak of amorphous calcite or amor-
phous aragonite. The proper term to use is amorphous calcium
carbonate.
Distinctive names, then, are necessary, or at least advisable,
for the amorphous equivalents of crystalline as well as for the other
amorphous minerals. As an illustration of the need of distinctive
names for amorphous minerals, let me cite the case of variscite. —
Cornu, in 1909, called its amorphous equivalent gelvariscite, but
Schaller has recently described lucinite, a dimorph of variscite.
_ Few new names are necessary, for varietal and other discarded
names may be used. In this paper I have recognized about twenty
tSee paper by the author, ‘““The Nomenclature of Minerals,’’ Trans. Am. Phil.
Soc., LIL (1913), 606-15.
2 Centralblatt f. Min. Geol. u. Pal., 1913, p. 68.
3 [bid., pp. 517-18.
524 AUSTIN F. ROGERS
of the best-defined amorphous minerals and have found it neces-
sary to introduce but two new names. In the future other new
names will be necessary, but their introduction will be gradual.
It is not desirable to recognize all amorphous mineral substances,
but only those that will stand the test of a critical examination, and
usually only those that are found in a number of localities. I
would emphasize especially the recognition of those with crystalline
equivalents, for then we have a comparison of properties which are
especially useful in determination.
SOME OF THE PROMINENT AMORPHOUS MINERALS
I now propose to describe and discuss what seem to be the
better established amorphous minerals, with especial emphasis
upon those that I have been able to study in more or less detail.
Amorphous carbon (schungite, mineral charcoal).—Graphite is
the hexagonal modification of carbon, and the so-called amorphous
graphite is probably compact, dense graphite in a fine state of.
division. Graphitite and graphitoid, according to Weinschenk,’
are simply varieties of graphite.
Schungite, described by JIostranzeff from near Schunga,
Government Olenez, Russia, is an amorphous modification of
carbon, for when treated with a mixture of potassium chlorate and
nitric acid it is soluble and is not, like graphite, converted into the
yellow, scaly substance called graphitic acid.
Amorphous sulfur (sulfurite).—Besides the common ortho-
rhombic, and the rare monoclinic, sulfur found at a few localities
amorphous sulfur also probably occurs in nature. Rinne’ has pro-
posed the name sulfurite for naturally occurring amorphous sulfur.
He describes an arsenical variety (arsensulfurite) which occurs as
amorphous crusts on andesite.
Xanthochroite CdS(H.O),? (greenockite in part)—The cad-
mium sulfid which occurs as a thin incrustation on sphalerite is
amorphous, as has been recognized by Lacroix and by Christo-
manos. It is usually called greenockite, but the original cadmium
sulfid first described from Scotland and found in but few other
« Zeit. f. Kryst. u. Min., XXVIII (1897), 291.
2 Centralblait f. Min. Geol. u. Pal., 1902, p. 499.
A REVIEW OF THE AMORPHOUS MINERALS 525
localities is hexagonal. A new name is necessary for amorphous
cadmium sulfid and I propose to call it xanthochroite (Greek xanthos,
yellow, chroa, color).
This mineral has recently been found. near Topaz, Mono
County, California, where it occurs as a thin coating on massive
magnetite. With the magnetite is associated sphalerite, and the
xanthochroite is doubtless a secondary mineral derived from
sphalerite. It varies from yellow to orange in color and is almost
opaque when examined with the microscope. ‘There is no evidence
of crystallization or double refraction. It is soluble in hydrochloric
acid and is reprecipitated by hydrogen sulfid. It is immediately
darkened by copper sulfate solution, and this will probably dis-
tinguish it from crystalline greenockite.
Hydrotroilite. FeS(H,O),.—The black slime found in inland
seas and in some moist sands and clays is colloidal and amorphous
ferrous sulfid. It has been described from Hadishibey Liman in
Southern Russia by Sidorenko' and the name hydrotroilite given to
it. As troilite is a synonym of pyrrhotite, the name is not a very
fortunate one, but it has priority. |
I am indebted to Mr. G. A. Waring, of the United States
Geological Survey, for a black slimy deposit from the Kruzgekampa
Spring, 60 miles north of Nome, Alaska. This consists of a black,
opaque substance mixed with sand grains and diatoms. The black
substance is hydrotroilite. It is soluble in cold hydrochloric acid
with the evolution of hydrogen sulfid. The solution gives tests for
ferrous iron, and with ammonia a black precipitate is obtained.
Wherry’ proposes to call the melnikowite of Doss? x-pyrite, but
melnikowite is microcrystalline (metacolloidal) FeS, and not amor-
phous. This is a serious objection to Wherry’s scheme of nomen-
clature.
Opal. SiO,(H.O), (lardite)—Opal is a typical amorphous
hydrogel and is unique in that, until Cornu’s work in 1909, it was
the only amorphous equivalent of a crystallized mineral generally
1 For reference to original article see Newes Jahrb. f. Min. Geol. u. Pal., II (1902),
ref. p. 397.
2 Central. f. Min. Geol. u. Pal., 1913, p. 518.
3 Neues Jahrb., Beil. Bd. XX XIII (1912), 689-93.
526 AUSTIN F. ROGERS
recognized as a distinct species. ‘The properties of opal are so
well known that a description is unnecessary. Attention, however,
should be called to the fact that hyalite opal, the purest and most
typical form of opal, usually shows double refraction due to strain.
(Lechateliérite)?>—The amorphous constituent of fulgurites
and of some inclusions in volcanic rocks has recently been named
lechateliérite by Lacroix... This material approaches silica glass in
composition and is also very similar to opal in properties. The
only difference between lechateliérite and opal is due to their
previous history, but, as Miers says, “. . . . the essential character
of a mineral, moreover, is quite independent of its source or previous
history.’ Yet these two substances are so different in occurrence
and origin that one feels inclined to consider them as distinctive
minerals. Then we are confronted with the question whether we
are ever to recognize more than one amorphous mineral for a
given crystalline equivalent or not.
Now, lechateliérite is a glass and may be considered along with
other natural glasses as a mineraloid (see p. 540) rather than as a
mineral proper. As an argument for this, I give the results of my
examination of a fulgurite found in the sand dunes along Lake
Michigan in Van Buren County, Michigan, and obtained from
Ward’s Natural Science Establishment. The fulgurite is a hollow
tube of glass with small grains of white sand adhering to the
exterior surface. The sand grains are quartz and orthoclase. The
glass is colorless and perfectly isotropic with an index of refraction
of 1.462+0.003. It is fusible on the edges and gives a small
amount of water in the closed tube. From these tests it can be
seen that the glass is not pure silica, but a glass high in silica.
The lechateliérite described by Lacroix is almost pure silica, as
its index of refraction is 1.458. Whether a distinctive name is
desirable or not, the glass of fulgurites may be considered simply
as a mineraloid, which approaches pure silica glass in composition.
Hydrocuprite. Cu,O(H2O), (hydrocuprite, ziegelite, ziguéline,
tile-ore).—Two kinds of cuprous oxid occur in nature: (1) iso-
metric cuprite and (2) amorphous tile-ore, for which the name
t Bull. Soc. Fran. de Min., XX XVIII (1915), 182-86.
2 Mineralogy, 1902, p. V.
A REVIEW OF THE AMORPHOUS MINERALS 527
hydrocuprite may be used. Werner considered rothkupfererz and
zeigelerz as co-ordinate. Beudant used the name ziguéline as a
species name for cuprous oxid. Hydrocuprite was described by
Genth’ as an orange-yellow to orange-red, amorphous raglike coat-
ing on magnetite from Cornwall. The same substance has been
noted by Lacroix on cuprite from Chessy, France, and by Sand-
berger, mixed with cuprite, from Schapbach, Baden. According
to Schaller? the supposed vanadium ocher from Lake Superior is
hydrocuprite (or possibly cuprite).
I have observed what I consider to be amorphous cuprous oxid
in specimens from the Lowell mine, Bisbee, Arizona; the Poderosa
mine, Collahuasi, Chile; and an unknown locality. The hydro-
cuprite occurs as a massive brick-red mineral associated with
cuprite. Under the microscope it appears as an orange-colored,
almost opaque substance in contrast to the dark-red translucent
cuprite. No very satisfactory optical tests can be made on account
of the opacity, but in a few spots the orange-colored mineral is
isotropic as well as the cuprite.
On account of the optically isotropic character of both of these
minerals the closed-tube test may prove useful. Hydrocuprite
contains water, but it is not a definite hydrate, as the name implies.
Melaconite. CuO(H.O), (tenorite in part, melanochalcite).—
Dana united the crystalline tenorite and amorphous melaconite
under the name tenorite, but they should be separated. Crystal-
line tenorite is a very rare mineral, known only from Vesuvius,
Cornwall, and Keweenaw Point, Michigan, but the amorphous
melaconite is a fairly common mineral in the oxidized zone of
copper mines. It is a black, massive mineral and occasionally
occurs in colloform crusts. In fragments it is black and opaque,
but is usually translucent brown and isotropic on the thin edges.
In addition to cupric oxid and water melaconite also contains silica,
the carbonate radical, and often manganese oxid.
Melanochalcite, described by Koenig as a copper salt of silico-
carbonic acid, is undoubtedly melaconite. Kraus and Hunts
t Preliminary Report on the Mineralogy of Pennsylvania, Pennsylvania Second
Geol. Surv., 1875, p. 46.
2Am. Jour. Sci. (4), XXXIX (1915), 404. 3-Tbid., (4), XLI (1915), 211-14.
528 : AUSTIN F. ROGERS
decide that melanochalcite is a mechanical mixture of tenorite,
malachite, and chrysocolla. While the two latter minerals may
sometimes be associated with it, homogeneous melaconite free from
mechanical impurities also gives tests for the carbonate radical and
silica. Melaconite is either an adsorption compound or a solid
solution of cupric oxid and water with silica, carbonate radical,
and often manganese dioxid.
Hematite. Fe.O,(H.O), (hydrohematite, turgite, hematogelite).
—Two varieties of ferric oxid are generally recognized, a crystalline
one and a massive or earthy-red one. Haiiy and, Werner respec-
tively considered these as distinct minerals under the names fer
oligiste, fer oxydé rouge and eisenglanz, rotheisenstein. ‘They were
united by Hausmann, and he has been generally followed by other
mineralogists. The red, earthy varieties of ferric oxid, such as the
odlitic Clinton ore and the soft hematites from the Mesabi range,
are amorphous and should be separated from the crystalline ferric
oxid. The amorphous ferric oxid may be distinguished from the
crystalline mineral by the fact that it contains a small amount of
water.
Turgite is usually considered to be 2Fe,0,°H.O, but it is prob-
ably not a definite hydrate. Turgite often occurs in colloform
crusts. It sometimes shows a fibrous structure, but this is prob-
ably because it is a dehydration product of limonite and retains its
structure. In fragments turgite is very dark red, sometimes iso-
tropic and sometimes birefringent. The double refraction may be
due to strain. Turgite is essentially identical with amorphous
hematite.
Specularite, now used as a varietal name by many mining
geologists, may be used as a specific name for crystalline ferric
oxid, and hematite then may be used exclusively for amorphous
ferric oxid.
Limonite. Fe,Hs0,(H.0),(?) (stilpnosiderite, melanosiderite,
limnite, esmeraldaite, xanthosiderite ?).—Leaving géthite out of
consideration, hydrous ferric oxid with a yellow-brown streak
occurs in two distinct forms, a crystalline fibrous form and an
amorphous massive form. Under the microscope the former ap-
pears as crystalline fibers with parallel extinction, and the latter is
A REVIEW OF THE AMORPHOUS MINERALS 529
optically isotropic and usually structureless. As Cornu suggests,
these two forms should have distinctive names. He used limonite
for the crystalline mineral and stilpnosiderite for the amorphous
one, but in view of the fact that the simpler name limonite is so
well established for the common and widely distributed brown
hydrated iron ore of surface origin it seems advisable to retain the
name limonite for the amorphous mineral. Limonite in this
restricted sense has priority over stilpnosiderite, for it was used by
Hausmann in 1813, while the name stilpnosiderite was introduced
by Ullmann in 1814 and usually has been used as a varietal name
for evsenpecherz.
A new name is necessary for the crystalline mineral with the
composition Fe,HsO,, but in view of the fact that the mineral
hydroxids of irod are being investigated by the Geophysical
Laboratory no suggestion is offered at present.
The melanosiderite of Genth is not an iron silicate, but limonite
with dissolved or adsorbed silica. ‘The water content of the amor-
phous ferric oxids is variable, and so it is probable that xanthosider-
ite, limnite, and esmeraldaite are not definite hydrates, but are all
simply varieties of limonite.
Stibiconite. Sb,O,(H,O), (stibianite, stiblith, stibioferrite, vol-
gerite)—The most common alteration product of stibnite is the
amorphous mineral stibiconite. It is probably colloidal antimony
tetroxid with adsorbed or dissolved water and not a definite hydrate.
The existence of crystalline antimony tetroxid is doubtful, for no
accurate description of cervantite has ever been made. Tenné and
Calderon state that much of the supposed Spanish cervantite is
valentinite.
Cliachite. Al,0;(H.O), (bauxite in part, wocheinite, sporogel-
ite, alumogel, shanyavskite).—Bauxite should be used as the name
of a certain type of rock and not as a mineralogical term. It was
so regarded by Dufrenoy, who introduced the name in 1845. He
has been followed by Tuéan, Tschermak, and Lacroix. Dittler and
Doelter, however, use bauxitite for the rock name.
The principal constituent of bauxite is an amorphous mineral
called cliachite (kliachite) by Cornu, who revived the name used by
Breithaupt.
530 AUSTIN F. ROGERS
The crystalline mineral corresponding to cliachite is hydrargil-
lite or gibbsite. This occurs to a minor extent in pisolitic bauxite
and also in some clays, but in the “granitic type’’ of bauxite from
central Arkansas’ hydrargillite is the principal mineral.
The composition of bauxite is usually given as Al,O,°2H,O,
but, as in the other amorphous minerals, the water content
of cliachite is variable. In the purest of the Georgia bauxites
the ratio of ALO, to H.O is very close to 1 to 3, according to
Watson.”
An amorphous mineral from India described by Warth3 under
the name gibbsite has the composition Al,O,°3H.O and is probably
cliachite.
Shanyavskite, an amorphous mineral with the composition
Al,0,°4H.O, is probably essentially the same as cliachite.
I have examined several specimens of cliachite and have found
the index of refraction to be 1.5700.005.
Pyrolusite. MnO,(H,0),.—Pyrolusite is probably an amor-
phous manganese dioxid corresponding to crystalline polianite.
The fibrous structure is due to the fact that it is pseudomorphous
after manganite. It always contains a small amount of water.
Psilomelane. Formula doubtful (wad, lithiophorite, asbolane,
lampadite).—Lacroix says that psilomelane is crystalline. There
may be crystalline equivalents of psilomelane, such as hollandite
and romanéchite, but most specimens of psilomelane show no indi-
cation of crystalline structure and are doubtless amorphous.
Psilomelane is probably a salt of some manganese acid and not
simply an oxid of manganese. Along with water it may contain
BaO, CaO, MgO, Fe,0,;, ALO;, K.0, Na.O, Li,O, CoO, and CuO.
A specimen from near Sodaville, Nevada, presented to me by
Mr. L. B. Spencer, is said to contain SnO, and WO,.
Collophane. 3Ca,;(PO,).-Ca(CO,,F.)(H-O), (apatite in part, col-
lophanite, fluocollophanite, quercyite, nauruite).—The principal
constituent of phosphorite or so-called phosphate rock is not mas-
sive apatite, but an amorphous substance which is identical with
t Mead, Econ. Geol., X (1915), 41.
2 Am. Geol., XXVIII (1901), 25.
3 Mineral. Mag., XIII (1902), 172.
A REVIEW OF THE AMORPHOUS MINERALS 531
the kollophan of Sandberger. Collophane’ is a calcium carbono-
phosphate approaching the formula given above. When the
mineral was first described, the calcium carbonate present was
thought to be an impurity and so was deducted from the analysis.
The chemical composition, as with most amorphous minerals, is
variable and somewhat uncertain, because it is difficult to dis-
tinguish mechanical impurities from essential constituents.
Fluorin is present in some specimens, and Lacroix uses the
name fluocolophanite for varieties rich in fluorin. Artini? describes
an amorphous mineral in the phosphorites of Es Salt, Palestine,
which he says is a fluorcollophanite near fluorapatite in composi-
tion. I have found that fossil bones consist of the mineral collo-
phane, and Carnot? has shown that the fluorin content of fossil
bones increases with the geological age of the bones. If there is an
amorphous equivalent of fluorapatite, should a distinctive name,
such as fluocollophane, be used for it? As more latitude must be
allowed in the chemical composition of amorphous minerals than
in the crystalline minerals, it hardly seems advisable to use more
than one name (collophane) for amorphous equivalents of minerals
of the apatite group. The fluorin determination is not an accurate
one, and fluorin in analyses of the phosphorites may be due to
fluorite or residual apatite.
Collophane is a massive mineral. and often has odlitic or con-
cretionary structure. In cavities it appears in colloform crusts
and often resembles opal. The color is sometimes white, but is
usually yellow, brown, or black, and is probably due to an organic
pigment.
Collophane is isotropic in part, but frequently shows double
refraction. In a specimen of phosphorite from Lassa Island the
double refraction of the colloform collophane is greater than that of
the accompanying fibrous dahllite. I have examined specimens
t Dana changed kollophan to collophanite, but since the name has not come into
general use it is preferable to use the English equivalent of the original, which is a
simpler and more euphonious name. Another argument is that collophanite may be
confused with colophonite, a variety of garnet.
2 Abstract in Zeit. f. Kryst. u. Min., LV (1915), 320.
3 Ann. de Mines (9), III (1893), 155.
532 AUSTIN F. ROGERS
of fossilized bones consisting of collophane which shows an appre-
ciable amount of double refraction. The double refraction is
variable from spot to spot, and the result is a characteristic wavy
extinction.
In view of these facts the safest method of distinguishing collo-
phane from dahllite is usually by means of the index of refraction.
I have determined the index of refraction of collophane from
many localities and have found that it usually varies from 1.57 to
about 1.61," while that of dahllite varies from about 1.61 to 1.63.
With high magnification dahllite shows a fabrous structure which
is lacking in collophane.
Like dahllite, collophane is soluble in hot nitric acid with
effervescence, but in the closed tube it gives a great deal of water
while dahllite gives little or none.
The quercyite of Lacroiz? is collophane and not a mixture of
dahllite and collophane. J have examined a specimen of a banded
calcium carbonophosphate mineral labeled ‘‘Hydroapatite, Mar-
seilles, France,” which is identical with Lacroix’s quercyite. The
doubly refracting layers of my specimen and probably of the
specimens figured by Lacroix (op. cit., p. 580) are collophane and
not dahllite. In support of this view it may be mentioned that
the index of refraction (1.598 0.002 in my specimen, but Lacroix
gives 1.608) is too low and the water content (3.2 to 6.0 per cent)
too high for dahllite. Double refraction is by no means a proof of
crystallinity. Lacroix also mentions the fact that the fibers of
quercyite lack the individuality of those of dahllite (and staffelite),
which is a good argument in favor of its being collophane.
Monite from Mona Island in the West Indies is also a variety of
collophane. It is a white, earthy, massive mineral which, under
the microscope, is largely isotropic with small doubly refracting
spots. The index of refraction is 1.631+0.001, a little higher
than for most specimens of collophane. Monite effervesces vig-
orously in hot nitric acid, so that the carbonate radical was over-
looked in Shepard’s analysis. . In the closed tube it gives abundant
t The fluorcollophanite described by Artini (loc. cit.) has an index of refraction of
1.630.
2 Mineralogie de la France, IV (1910), 579.
A REVIEW OF THE AMORPHOUS MINERALS 533
water. Monite also contains the sulfate radical, but this is not
present as gypsum, as Shepard thought, for the mineral is homo-
geneous except for minute crystals of monetite and colloform crusts
of dahllite. The sulfate radical probably replaces part of the
carbonate radical. Phosphorites from Idaho and South Carolina
also contain the sulfate radical.
Nauruite described by Elschner’ from the island of Nauru is
another synonym of collophane. It is a yellow to brown resinous
mineral occurring as agate-like layers in cavities of phosphorite.
Snes . Theanalysesof the
Nauru phosphorites all show calcium carbonate, and specimens of
typical nauruite kindly furnished by Mr. Elschner effervesce
vigorously in hot nitric acid. Under the microscope the Nauru
collophane shows banded spherulitic structure and weak double
refraction. The index of refraction is 1.5970.001, which proves
that the mineral is not dahllite in spite of its birefringence. The
double refraction is variable and is lost upon heating.
I have examined thin sections of phosphorites or so-called phos-
phate rocks from Florida, South Carolina, Tennessee, Kentucky,
Idaho, Nauru, Fanning Island, Ocean Island, Lassa Island, and
Clarendon, New Zealand, and have found amorphous collophane
‘to be the principal constituent of all of them. It is often accom-
panied by dahllite. It seems strange that so little attention has
been paid to these carbonophosphate minerals, for they are impor-
tant from both the economic and scientific standpoints.
Evansite. Al,(OH)6PO,- (H.O), ?—Evansite, a hydrous alumi-
num phosphate described by Forbes in 1864, is a typical amorphous
mineral. It occurs asa colloform incrustation and greatly resembles
allophane. ‘The index of refraction determined on a specimen from
Zelegnik, Hungary, is 1.483+0.003.
Evansite has recently been described from two American locali-
ties (Goldburg, Idaho, and Columbiana, Alabama) by Schaller.?
Pitticite. FeAsO,'Fe,O;:(H.O), ?—Pitticite is a dark-brown
massive mineral resembling limonite. It is a basic ferric aresenate
The formula given is 3Ca,;P,03—
t Corallogene Phos phat-Inseln Austral Oceanien und Ihre Produkte (Liibeck, 1913),
Dp. 54.
2 Bull. 490, U.S. Geol. Surv., 1911, p. 94.
534 AUSTIN F. ROGERS
and often contains the sulfate radical. An artificial colloidal basic
ferric arsenate has been described by Holmes and Rundfusz.*
I have examined specimens of pitticite from two localities, one
on the west side of the South Merced River in Mariposa County,
California, and the other at Sioux City, Iowa. At the first locality
it is an oxidation product of arsenopyrite and at the second it
occurs with limonite and aragonite.
Pitchblende UO.,UO,,(H.O), (Uraninite in part, nasturan).—
The colloform or massive pitchblende seems to be the amorphous
equivalent of isometric uraninite. Its water content is higher and
its nitrogen and rare earth content lower than that of uraninite.
It may also be distinguished by its lower specific gravity (6.5 to 8;
uraninite is 9 or above).
Uraninite occurs in pegmatites while pitch piende occurs in
metalliferous veins with sulfids.
(Maskelynite) >—Maskelynite was first described by Tschermak
from the Shergotty (India) meteorite. It is optically isotropic and
in composition is very close to a plagioclase with equal molecular
percentages of albite and anorthite. As the index of refraction
is almost exactly the same as that of an artificial glass’ of the compo-
sition Ab,An,, it is probably the amorphous equivalent of plagioclase
(not necessarily a fused plagioclase) and not an isometric mineral.
Maskelynite is on the same footing as lechateliérite. They are
the only glasses which have been regarded as minerals within recent
years. They may be disposed of by considering them mineraloids
rather than as definite minerals.
Malacon. ZrSiO,(H.O),? (cyrtolite, auerbachite, oerstedite,
ostranite).—Malacon is the name given to a mineral from Hitter6,
Norway, which has the form of zircon, but is softer and contains
water. It is undoubtedly an alteration product and in all probabil-
ity the amorphous equivalent of zircon. It deserves recognition
as a distinct mineral, and Lacroix so regards it.
Malacon may be distinguished from zircon, not only by its
isotropic character and water content, but also by its inferior hard-
ness and specific gravity and its lower index refraction (n= 1.826
1 Jour. Am. Chem. Soc., XX XVIII (October, 1916), 1970.
2 Larsen, Am. Jour. Sci. (4), XXVIII (1909), 267.
A REVIEW OF THE AMORPHOUS MINERALS 535
Larsen). The other minerals mentioned above (cyrtolite, etc.) are
probably synonyms of malacon.
Halloysite. ,A1.Si,0,(H.O), (myeline, indianaite, clayite, bole
in part, kaolin in part).—It is generally recognized that some
of the clays consist in whole or in part of amorphous aluminum
silicates. Mellor’ has proposed the name clayite for the amorphous
constituent, but fortunately halloysite has priority.
Halloysite may be regarded as the amorphous equivalent of
crystalline kaolinite, for it has the formula H,ALSi.0,+,(H.O.) In
LeChatelier’s experiments on halloysite the water given off below
250 C. varied within wide limits, while the water given off above
this temperature was fairly constant and corresponded to that of
kaolinite.
Myeline is an indurated clay from Rochlitz, Saxony, which is
almost identical with kaolinite in chemical composition. I have
examined a specimen from the type locality and find it to be
essentially amorphous (there are patches of micro-crystalline
material) with an index of refraction of 1.5560.001. Gagel* has
described a non-crystalline myeline-like substance which occurs as
an alteration product of trachydolerite at Canical, Madeira.
I have examined indianaite from St. Lawrence County, Indiana,
which the published analyses show is a very pure halloysite,
and find it to be an amorphous mineral with an index of
refraction of 1.5380.002, which is less than that of kaolinite
(n=1.567—1.561).
A clay from Morton, Minnesota, with imperfect pisolitic struc-
ture consists largely of an amorphous substance with an index of
refraction of 1.557 0.003, and hence may be referred to halloysite.
This clay also contains some hydrargillite.
I have examined “kaolin”’ from Broken Hill, New South Wales,
and find it to be amorphous with m=1.548+0.002. This must
be halloysite.
The “isotropic kaolinite-like mineral” described by Larsen and
Wells? from Wagon Wheel Gap, Colorado, has an index of refraction
t Trans. Eng. Cer. Soc., VII (1908).
2 Central. f. Min., Geol., u. Pal., 1910, p. 225.
3 Proc. Nat. Acad. Sci., II (1916), 364.
536 : AUSTIN F. ROGERS
of 1.557 and a water content of 15.5 per cent, and hence should be
referrred to halloysite.
Helmhacker' describes halloysite with a small botryoidal sur-
face from Banat, Hungary. The Rochlitz myeline also shows
colloform crusts in cavities.
From the available evidence it seems clear that halloysite is
the amorphous equivalent of kaolinite. The recognition of this
may help in a clearer understanding of the constitution of clays.
The determination of the index refraction, which varies from 1.538
to 1.557, is probably a safer means of distinguishing halloysite
from kaolinite than the absence of double refraction. The pres-
ence of aluminum hydrate in solid solution may, however, modify
the index of refraction.
Allophane. Al,SiO;(H.O), (carolathine).—Allophane has long
been recognized as one of the typical amorphous minerals. Its
colloform shape, optical isotropism, and absence of cleavage
and structure leave no doubt as to its amorphous character.
The index of refraction of a specimen from Bedford, Indiana, is
1.473+0.002.
Whether schrotterite and collyrite should be considered syno-
nyms of allophane is open to question.
Stevensite. H.,Mg,(SiO,),°(H.O), (talc in part, lucianite ?).—
An alteration product of pectolite found at several localities in
New Jersey is probably the amorphous equivalent of crystalline
talc. This substance was first described by Leeds? who used
stevensite as a name for talc pseudomorphous after pectolite.
Glenn has recently studied this mineral, and his work proves that
it should be considered a distinct mineral. Wherry made a
microscopic examination and shows that it is isotropic with an
index of refraction of about 1.50. Stevensite is also distinguished °
from talc by its lower specific gravity and higher water content.
Stevensite is probably not a monohydrate of talc, as Glenn
suggests, but the water content is evidently variable and reaches
over 19 per cent in the clay mentioned in the next paragraph.
1 Min. u. petr. Mitth., IL (1879), 232.
2Am. Jour. Sct. (3), VI (1873), 22-23.
3 American Mineralogist, I (1916), 44-46.
A REVIEW OF THE AMORPHOUS MINERALS 537
Hilgard’ has recently described, from a locality near the City
of Mexico, a peculiar clay which consists largely of a hydrous
magnesium silicate. This substance was named lucianite, but as it
is a colloidal substance with specific gravity of 2.25 and is soluble in
hydrochloric acid it is probably a synonym of stevensite.
Cornuite. mCuO°nSiO.*(H.O), (chrysocolla in part).—In spe-
cimens of chrysocolla from a number of localities I have noted a
glassy, green or bluish-green copper silicate which is the amorphous
equivalent of chrysocolla. To this newly recognized mineral I
wish to apply the name cornuzte in honor of the late Dr. Felix Cornu,
of Leoben, Austria, who was practically the first mineralogist to
make a sharp distinction between crystalline minerals and their
amorphous equivalents.
I have found this mineral on specimens from Globe, Arizona;
Bisbee, Arizona; Ludwig, Nevada; Copper Mountain, Alaska,
and Collahuasi, Chile. It is usually optically isotropic with
varying from 1.525 to 1.549, but sometimes has irregular, weak
double refraction and wavy extinction. It is associated with
crystalline chrysocolla and often appears in colloform bands within
layers of colloform chrysocolla. It is more readily soluble in
hydrochloric acid than chrysocolla and also somewhat softer.
The best specimens of cornuite in my possession are from the
mine of the Alaska Consolidated Mining and Smelting Company at
Copper Mountain, Prince of Wales Island, Alaska. It occurs
as a beautiful bluish-green (Ridgway 42k), transparent, glassy,
somewhat banded crust about 1 cm. thick associated with chryso-
colla. The index of refraction is 1.549+0.001. Jam indebted to
my collegue, Dr. G. S. Bohart of the chemistry department, for
the following analysis of the Copper Mountain cornuite:
Ratios
CuOv 420m eR we 0.537
AL OF) OBIS dei. year oe 0.004
SIO3 S33 Ae 03 inte ae esos 0. 566
fa GSS as oh ee al ee tl 1.284
These figures are each the average of two closely agreeing values
made upon carefully selected material free from the associated
t Proc. Nat. Acad. Sci., II (1916), 8-12.
538 AUSTIN F. ROGERS
chrysocolla and all other impurities. The ratios seem to show an
excess of both silica and water over that required for the usually
accepted empirical formula (H,CuSiO,) for chrysocolla, which
probably indicates that cornuite is a solid solution of cupric oxid,
silica, and water.
Probert’ describes a “jelly of the most beautiful shades of
blue. and green” with the following composition: CuO=47. 46;
S10,= 21.20; H,O=28.05; CaO=1.39; AlLO;=tr., from the
200-foot level of Ray Central mine at Ray, Arizona. Here we
evidently have cornuite in process of formation.
Chrysocolla is sometimes considered to be an amorphous mineral,
but while chrysocolla shows colloidal structures it is microcrystal-
line (metacolloid), and recently Umpleby’ has described crystallized
chrysocolla from Mackay, Idaho, the indices of refraction of which
are W,=1.57 and 1,=1.406.
OTHER POSSIBLE VALID AMORPHOUS MINERAL SPECIES
The twenty-three amorphous minerals described in this paper
are all believed to be well-established mineral species. Practically
all of them are known from several localities and some of them are
very common and widely distributed minerals. Although their
properties vary somewhat, these minerals are fairly definite and can
be recognized by careful work.
There are, however, as many more amorphous minerals which
have been described or named, and some of these it may be possible
to establish by study of suitable material. I have attempted to
enumerate here some of the probable amorphous minerals.
Metastibnite, amorphous Sb:S;.
Jordisite, colloidal MoS,.
Patronite, vanadium sulfid. :
Ostwaldite (Buttermilcherz), colloidal AgCl.
Ehrenwerthite, colloidal Fe.0;°H.O.
Amorphous equivalent of fischerite.
Schadeite, amorphous equivalent of plumbogumnite.
Palmerite, hydrous aluminum potassium phosphate.
Yukonite, hydrous calcium iron arsenate.
1 Min. and Sci. Press, CXII (1916), 808.
2 Jour. Wash. Acad. Sci., IV (1914), 181.
A REVIEW OF THE AMORPHOUS MINERALS 539
Gummite, alteration product of uraninite.
Glockerite, hydrous ferric sulfate.
Plumboniobite.
Amorphous equivalent of thorite.
Amorphous equivalent of allanite.
Amorphous equivalent of gadolinite.
Amorphous equivalent of yttrotantalite.
Amorphous equivalent of homilite.
Amorphous equivalent of yttrocrasite.
Amorphous equivalent of pyrophyllite.
. Greenalite, hydrous ferrous silicate.
Yttrialite.
Neotocite.
Plombierite.
Geolyte (Bodenzeolith), amorphous equivalent of zeolites.
Pochite.
MINERALOIDS (HYDROCARBONS AND GLASSES)
In addition to the foregoing definite amorphous minerals there
are several other classes of naturally occurring amorphous sub-
stances. I refer to the hydrocarbons and glasses. Shall they be
considered as minerals or not? The answer depends upon our
definition of the term mineral or mineral species. A mineral
(species) is usually defined as a naturally occurring homogeneous
substance of definite chemical composition. By common consent
the term is limited to the naturally occurring substances, although
the specific mineral name is often used for the corresponding arti-
ficial substance. No objection can be raised if the word synthetic
or artificial is prefixed to the mineral name. In view of the dis-
covery of solid solutions of a kind different from isomorphous mix-
tures in minerals, the definition given above must be modified
so as to read “of more or less definite chemical composition,”’ as
suggested by Wherry."
In the case of the hydrocarbons the principal objection against
considering them minerals is on account of their organic character.
While directly or indirectly the result of organic growth, they are
on a somewhat different footing from ordinary plant products.
They occur with other minerals in sedimentary rocks and are, in
t Ibid. pp. 111-14.
540 AUSTIN F. ROGERS
fact, fossil resins and fossil fuels. They are collected and studied
by geologists.
The hydrocarbons are by some mineralogists given full rank as
minerals, by others omitted entirely, by still others treated in an
appendix to minerals. The latter procedure seems to be the safest
plan, for they are organic substances, and the typical minerals are
certainly inorganic, yet on the other hand they deserve some recog-
nition by the mineralogist. The mineralogist is often called upon
to identify them, and they should be described from a mineralogical
standpoint.
For these reasons I think the hydrocarbons may be included
under Niedzwiedzki’s term mineraloid.' As Niedzwiedzki used
this term for all naturally occurring amorphous substances, this
changes somewhat the original definition of mineraloid. Such
substances as opal, cliachite, limonite, collophane, halloysite, etc.,
are definite enough to be called minerals even though they are
amorphous. The term ‘“‘mineraloid’”’ seems appropriate for the
less definite mineral-like substances.
And in the same category I would also place the natural glasses.
As far as I know, glass is not given a place in any modern mineral-
ogical treatise, although the older mineralogists described tachylyte
and hyalomelane as mineral species. It is, however, included in
some of the determinative tables of the rock-forming minerals
found in textbooks on petrography, and for the same reason that it
is included in these tables it may be treated as a mineraloid.
Natural glass is a homogeneous substance to be identified the same
as minerals in general.
Natural glass is, of course, varied in composition in comparison
with the various types of igneous rocks, yet the average obsidian is
probably not much more varied than some of the amorphous miner-
als. While glass is scarcely entitled to recognition as a mineral,
there are arguments in favor of classifying it as a mineraloid. It
may well receive a place in the appendixes of our books on miner-
alogy. This will call attention to its properties and will aid in its
identification, which otherwise might be difficult for a beginner who
had studied mineralogy but not petrography.
t Centralblatt f. Min., Geol., u. Pal., 1909, pp. 661-63.
A REVIEW OF THE AMORPHOUS MINERALS 541
Glass, like amorphous substances in general, is in part optically
isotropic and in part birefringent. The birefringence is shown in
many perlites, some varieties of which are veritable Prince Rupert’s
drops. Besides the chemical composition, the most important
property of glass is its refractive index, which, as has been shown by
Stark,* varies from about 1.485 for ‘“‘acid”’ glasses with 75 per
cent silica up to 1.67 for ‘“‘basic”’ glasses with 4o per cent silica.
Obsidian, perlite, pitchstone, etc., are petrographic terms, but
glass as a whole may be treated as a mineraloid from the mineral-
ogical standpoint. Lechateliérite and maskelynite are glasses
which are fairly definite in chemical composition.
SUMMARY
Some of the naturally occurring amorphous substances are
definite enough to be recognized as mineral species.
The amorphous equivalents of crystalline minerals should be
treated as distinctive minerals and should have distinctive names.
About twenty of the more prominent and well-defined amor-
phous minerals are described and discussed. Most of these minerals
are the amorphous equivalents of crystalline minerals.
New names are given to amorphous cadmium sulfid (xantho-
chroite) and amorphous copper silicate (cornuite) which corre-
spond to sphalerite, greenockite, and chrysocolla respectively.
Arguments are advanced for treating the natural hydrocarbons
and natural glasses as mineraloids.
t Min. u. Petr. Mitth., XXIII (1904), 536-so.
THE CHAMPLAIN SEA IN THE LAKE ONTARIO BASIN
KIRTLEY F. MATHER
Queen’s University, Kingston, Canada
INTRODUCTION
The retreating front of the Labradorian ice sheet late in the
Wisconsin glacial stage rested for a time against the northwestern
slopes of the Adirondack Mountains. Between the ice front in the
upper St. Lawrence Valley and the southern rim of the Ontario
basin in New York state, the waters of Lake Iroquois were ponded.
The history of this lake, with its outlet down the Mohawk Valley
past Rome, New York, has been deciphered by Fairchild, Taylor,
Spencer, Coleman, and others. Further withdrawal of the ice
margin permitted the escape of the Iroquois waters through
‘Covey Gulf,” southwest of the summit of Covey Hill, the north-
ernmost hill on the west flanks of the Adirondacks, a mile north
of the international boundary. The altitude of the Covey outlet
at the present time is about 1,000 feet, while that of the Rome out-
let is 460 feet, but, according to Fairchild, the altitudes of the two
outlets during Iroquois time were very similar, ‘if not practically
identical” (see Fig. 1).
North from Covey Hill the land drops rapidly away to the broad,
low valley of the St. Lawrence. As soon, therefore, as the edge of
the ice sheet had withdrawn a mile or two farther northward, Lake
Iroquois was drained. The water in the Ontario basin and the
St. Lawrence Valley rapidly fell to sea-level, for the land stood at
a much lower altitude then than now. Differential uplift of the
Great Lakes region had commenced long before the extinction
of Lake Iroquois, and it is commonly held that the Champlain Sea
was then at its maximum extent. It has been more or less uncon-
sciously assumed that the history of these sea-level waters in the
upper St. Lawrence Valley was simply a slow but progressive
tH. L. Fairchild, “Pleistocene Uplift of New York,” Geol. Soc. Amer. Bull.,
XXVII (1916), 235-62.
542
THE CHAMPLAIN SEA IN THE LAKE ONTARIO BASIN 543
withdrawal from the maximum of submergence to the present maxi-
mum of emergence. Data now available, however, lead to the
conclusions that, when the ice barrier was removed from the St.
RAR a a deen SUE rit A)
mee WAVED TELE EE Ty y dis da RAD S
st (AULT 1] sd 1) ORIAN ICE
! / . J
\\ | VAS = Ay) Pett
/ f
=
VY 7%
7 he
aA O rer hay
be Pea Press vi
/ \
Tia gain that
ONTARIO
\ Ue
HL yy Vtg
0
ADIRONDACK
MOUNTAINS
LAHE /ROQUOIS
LAKE 1ROQUO/S
AT THE TIME THE
COVEY OUTLET
WAS ESTABLISHED
ca/e
o iS 30 4F 60 i
ly
Miles
PRESENT ALTITUDE OF SHoRE
LIné (NOIC ATED IN FEET
Fic. 1.—A late stage in the history of Lake Iroquois. Lake Algonquin, a portion
of which is shown in the Georgian Bay region, overflows through the Fenelon Falls
outlet east of Kirkfield, Ontario, and Algonquin River carries its waters to the Rice
Lake embayment of Lake Iroquois. The Niagara River outflow from Lake Erie also
contributes to Lake Iroquois, which is indicated in its two-outlet stage. Part of its
water spills into the Mohawk Valley near Rome, New York, and part falls over the
cliff at Covey Gulf, Quebec, into the marine embayment in Champlain Valley. St.
Lawrence and Ottawa rivers and parts of the present Great Lakes are indicated by
dotted lines. Pleistocene geography based largely upon the work of Fairchild, Cole-
man, Johnston, and Taylor. 4
Lawrence Valley north of Covey Hill, the level of the Champlain
Sea was far below its maximum height, and that the strand line
moved gradually up the valley of the St. Lawrence River and its
544 KIRTLEY F. MATHER
tributaries to the position of greatest submergence (Fig. 2) before
it began to withdraw toward its present location.
ligarsg:)
ADIRONDACK ~ y™
MOUNTAINS
CHAMPLAIN SEA
AT THE TIME OF
MAXIMUM SUBMER GENCE
cole
oO 1 3% 45 60 7
ae ee COR?
Miles
Fic. 2.—The geography of the Ontario-St. Lawrence Valley at the close of Pleisto-
cene time. Champlain Sea is at its maximum extent. Lake Algonquin has been re-
placed by the Nipissing Great Lakes, which overflow northward down Mattawa and
Ottawa rivers. Trent River is a comparatively small stream heading in the Trent
chain of lakes and emptying into Gilbert Gulf above Trenton, Ontario. Paleogeog-
raphy of New York state after Fairchild. Boundaries in Ontario and Quebec gener-
alized and only approximately correct because of scarcity of exact data.
EVIDENCE OF A PROGRESSIVE MARINE SUBMERGENCE OF THE
ONTARIO BASIN
Napanee Valley..—Napanee River is one of the many south-
westerly flowing streams tributary to Lake Ontario along its
tT am indebted to N. B. Davis, of the Department of Mines, Ottawa, for directing
my attention to the surficial deposits in Napanee Valley as well as for valuable sug-
gestions made in the field in September, 1916.
THE CHAMPLAIN SEA IN THE LAKE ONTARIO BASIN 545
northern shore. ‘The river heads in a chain of little lakes which
dot the surface of the pre-Cambrian rocks in the central part of
Frontenac County, 25 miles north of the east end of Lake Ontario.
The lowest of the small lakes is Lake Napanee, 446 feet above sea-
level. Thence the river flows across the nearly flat-lying Ordovi-
cian limestones which overlap the pre-Cambrian complex, past the
towns of Yarker and Napanee to the Bay of Quinte on Lake
Ontario (see Fig. 3). From Lake Napanee to Lake Ontario the
length of the river is about 24 miles, and its total fall is 200 feet.
Throughout this portion of its course Napanee River occupies a
pre-Wisconsin youthful valley cut in the limestone cuesta and con-
sequent upon its slope. The valley floor is 75-125 feet below the
intervalley upland surface and is bounded by abrupt limestone
scarps. Near Yarker the valley walls are scarcely a quarter of a
mile apart, but toward Napanee they diverge gradually to a dis-
tance of a mile and a half.
The upland surface on either side of Napanee Valley displays
abundant evidence of wave and current action. Bedded clays in
many places provide a thin veneer of fertile soil over the barren
glaciated limestone surface. Elsewhere, bared rock surfaces are
studded with the larger bowlders from glacial drift; all the finer
products of the glacial mill have been washed away by currents
and waves. No distinctive shore features have as yet been
observed along the Napanee Valley to mark the upper limit
of wave action, but marine or lacustrine clays are present up
to altitudes of at least 450 feet in the vicinity of Yarker. At
Inverary, 15 miles due east from Yarker and 11 miles north from
Kingston, M. B. Baker and I have definitely located the limit
of wave action at elevations between 500 and 510 feet above sea-
level.
The region is within the area covered by the ice barrier to which
Lake Iroquois owed its existence and could not have been freed
from its glacial burden until after the extinction of that lake.
The wave and current action is therefore that of Gilbert Gulf,
the portion of Champlain Sea which occupied the Ontario
basin after removal of the ice dam from the Thousand Island
region.
546 KIRTLEY F. MATHER
Fic. 3.—Index map of the Kingston-Napanee district, Canada, along the northern
shore of the eastern end of Lake Ontario.
THE CHAMPLAIN SEA IN THE LAKE ONTARIO BASIN 547
Near Ottawa the upper limit of marine submergence has recently
been determined’ to be 690 feet above sea-level. At Belleville the
highest marine plane has an altitude of 323 feet.?__ If the gradient
is regular between those two localities, the altitude at Yarker would
be close to 440 feet. The nearest point on the shore line of Gilbert
Gulf in New York, where its features have been mapped by Fair-
child, is a small hill. 45 miles southwest from Clayton. This hill
is 35 miles S. 70° E. from Yarker; the Gilbert shore encircles it at
an elevation slightly above 400 feet. The post-Champlain isobases
in this neighborhood run approximately east and west. It is,
therefore, quite clear that the Napanee Valley and adjacent uplands
were submerged beneath the waters of Gilbert Gulf at least as far
northward as Yarker (elevation 425 feet).
That this submergence did not take place when the ice front
stood south of the divide at the head of Napanee Valley is
clearly indicated by the presence of a valley train of fluvio-glacial
gravels within the valley. Remnants of the gravel beds are pres-
ent at many localities between Yarker and Napanee. Between
Newburgh and Strathcona their fluvio-glacial, rather than glacio-
lacustrine or glacio-marine, origin is readily apparent. As indi-
cated in Fig. 4, the gravel train has a plane upper surface and its
remnants are now distinctly terrace-like. The summit of the
valley train is approximately 50 feet above the modern valley flat
and has a similar gradient, 8 or to feet to the mile. The gravels
and sands are irregularly bedded; cross-bedding is common, and
assortment according to size of pebbles is very incomplete. Many
of the stones are striated or faceted by glacial action. The valley
train is in every way a typical sub-aerial fluvio-glacial deposit similar
to the comparable outwash gravels of the Fox River Valleyin Illinois.
The lowest and most southwesterly remnant of the Napanee
valley train is within the city limits of Napanee at an elevation of
tW. A. Johnston, ‘‘Late Pleistocene Oscillations of Sea-Level in the Ottawa
Valley,” Canada Geol. Survey, Mus. Bull. 24, 1916, p. 5.
2F. B. Taylor, ‘“‘Gilbert Gulf,” U.S. Geol. Survey Mon. 53, 1915, Pp. 445-46;
A. P. Coleman, ‘‘Marine and Freshwater Beaches of Ontario,” Geol. Soc. Amer. Bull.,
XII (1901), 129-46.
3H. L. Fairchild, ‘Pleistocene Features; Clayton-Lafargeville District,’ NV.Y.
State Mus. Bull. 145, 1910, pl. 46.
548 KIRTLEY F. MATHER
less than 325 feet. In a small gravel pit beside the east abut-
ment of the stone railway bridge the gravels and sands are
overlain by bedded clays. Similar bedded clays are present
on the floor of the valley eroded into the fluvio-glacial gravel
deposits.
At the time of construction of the Napanee valley train the
retreating front of the ice sheet had withdrawn less than 10 miles
north of Yarker, and the marine plane could not have been over
Fic. 4.—Napanee Valley between Newburgh and Strathcona, Ontario. The
cattle in the foreground and the houses and barns in the distance are on the surface
of the Napanee valley train. The lower flat in the middle distance is the modern flood-
plain of Napanee River. The exposure of fluvio-glacial gravels and sands in the escarp-
ment overlooking the valley flat is in part due to gravel pits and in part to the railroad
which parallels the river at the foot of the scarp. Beyond the buildings in the distance,
the land rises abruptly to the summit of the limestone cuesta which forms the sky line
at the left.
325 feet above present sea-level. Further retreat of the ice cut off
the supply of gravel, and dissection of the valley train commenced.
Subsequently sea-level waters crept upward and submerged the
whole region to at least the 425-foot contour line. In these waters
bedded clays accumulated.
Trent Valley—The anomalous relation between the Trent
Valley spillway from Lake Algonquin and the Gilbert Gulf shore
features has long been a puzzle to glacialists. ‘The physiographic
THE CHAMPLAIN SEA IN THE LAKE ONTARIO BASIN -549
. features have been described recently by Taylor* and Johnston?
and need not be dwelt upon here. Briefly, Algonquin River carried
the overflow from Lake Algonquin to Lake Iroquis and, after its
extinction, to Gilbert Gulf. A large delta in the Rice Lake region
marks the point at which the river debouched into Lake Iroquois.
It is now approximately 620 feet above sea-level. The spillway
channel continues down Trent Valley past the Gilbert Gulf shore
line (320 feet) to the present level of Lake Ontario (246 feet) near
Trenton. The channel below the summit marine plane apparently
differs in no way from the portion above that level.
To explain these features by postulating a continuance of the
Algonquin River flow until after the lower portion of Trent Valley
had been lifted above the level of Champlain Sea is obviously
difficult. Neither Taylor nor Johnston is satisfied with that
explanation. In the light of the conclusions resulting from the
study of Napanee Valley, the difficulties met with in Trent Valley
are removed. Algonquin River carried its large volume of water
to, and below, the present level of Lake Ontario before, rather than
after, the Trenton neighborhood was submerged beneath sea-level
waters. The delta which must have been built at the outlet of
the river into Gilbert Gulf is now hidden beneath the lake. Sound-
ings in the Bay of Quinte may some time reveal its hiding-place.
By the time the Gilbert strand had crept above the present lake-
level retreat of the ice in the Nipissing region had uncovered a new
outlet for the upper Great Lakes, and the Lake Algonquin overflow
was diverted from the Fenelon Falls outlet.
Confirmatory evidence.—Physiographic features in Napanee and
Trent valleys are thus explainable by postulating a positive move-
ment of the strand line in the Ontario basin during the final stages
of the waning ice sheet. A similar advance of marine waters in
Ottawa Valley has been suggested by Johnston’ to explain the.
tF. B. Taylor, ‘‘The Pleistocene of Indiana and Michigan and the History of
the Great Lakes” (Leverett and Taylor), U.S. Geol. Survey Mon. 53, 1915, pp. 445-40.
2W. A. Johnston, ‘‘The Trent Valley Outlet of Lake Algonquin and the De_
formation of the Algonquin Water-Plane in Lake Simcoe District, Ontario,” Canada
Geol. Survey Mus. Bull. 23, 1910.
3W. A. Johnston, ‘‘Late Pleistocene Oscillations of Sea-Level in the Ottawa
Valley,” Canada Geol. Survey Mus. Bull. 24, 1916.
550 KIRTLEY F. MATHER
relations between the fossil-bearing clay zones in the vicinity of
Ottawa.
The quaternary deposits near Waterville, Maine, suggest a
similar movement of the sea-level there. H. P. Little’ states that
‘the main mass of the fluvio-glacial deposits is found in an esker
. . . . bordered by marine clays and sands. These overlap the
esker and are separated from its gravels by an unconformity con-
sidered due to sub-aerial erosion.”’
CHANGES IN SEA-LEVEL INCIDENT UPON THE WASTING OF THE
LABRADORIAN ICE SHEET
The suggestion is an obvious one that the positive movement
of the strand line in late Pleistocene times in Northwestern North
America may be a result of the return of water to the ocean basins
from wasting ice caps. The simple effect of ice melting is, however,
complicated in the region under discussion by close proximity to the
ice masses. Here, too, sea transgression has been followed by a
presumably much greater negative movement of the strand, which
has continued nearly or quite to the present time. Barrell’s?
discussion of the various problems involved is especially stimulating
in this connection.
Three factors enter into the local problem: elevation of sea-
level due to return of water which had been congealed on the land;
depression of sea-level due to decreasing gravitative attraction of
the ice masses; and uplift of the land due to isostatic, or other,
readjustment consequent upon removal of ice burden. None of
the three can be exactly evaluated from the data now available,
and the effect of the third can be estimated only from field evidence.
Mathematical calculations will, however, help to crystallize opinion
concerning their interaction.
Woodward’s classic contribution’ to the subject does not exactly
meet the situation at hand. His assumptions concerning area and
1H. P. Little, ‘‘Pleistocene and Post-Pleistocene Geology of Waterville, Maine,”
abstract of paper presented before the Geological Society of America, December, 1916.
2 J. Barrell, ‘‘Factors in Movements of the Strand Line and Their Results in the
Pleistocene and Post-Pleistocene,” Am. Jour. Sci. (4), XL (1915), 1-22.
3R. S. Woodward, ‘‘On the Form and Position of the Sea-Level,” U.S. Geol.
Survey Bull. 48, 1888.
THE CHAMPLAIN SEA IN THE LAKE ONTARIO BASIN 551
thickness of the ice cap were made with the expressed purpose’ of
determining the ‘‘maximum upheaval of the water’’ possible as a
result of glacial conditions. I have, therefore, with indispensable
assistance from my wife, computed the results obtained from his
formulas on the basis of quite different assumptions, which are
believed to be more in accord with known data concerning the
Labradorian ice sheet.
For mathematical purposes it 1s fair to assume a circular ice
cap with a radius of 14°, or about 966 miles, on the earth’s surface
and a thickness at its center of 5,000 feet. Let its surface configura-
tion correspond to 7=10 in Woodward’s equation 95.?__ This would
give an increasingly rapid slope of the ice surface from center to
border as in the following table:
Slope in Feet per Mile of Ice Surface at Varying Distances, Expressed in
Degrees, from Center of Assumed Ice Cap
Center. 4 6° 8° 10° 12° Border
From equation 96 it follows that such an ice cap would lower
the sea-level 82.5 feet, if all its ice were formed from moisture
withdrawn from the ocean. Following the method of Woodward in
§ 49,3 we find that the gravitative attraction of this ice sheet would
distort the level of the sea so that the disturbed surface along the
border of the ice mass would be 136 feet above the undisturbed
surface. The average slope of the disturbed surface within one
degree of the ice border would be 0.13 feet per mile. If the ice
cap were 10,000 feet in thickness at its center, the distortional effect
would be twice as great as that of a 5,000-foot cap of the same
diameter and surface contour.
There is every reason for believing that the development of
the half-dozen ice sheets which covered parts of North America
and Europe in Pleistocene times was practically synchronous.
The total effect of all must, therefore, be considered in evaluating
Ota. DIES:
2 Tbid., p. 62. 3 [bid., pp. 65-66.
552 KIRTLEY F. MATHER
resultant changes in sea-level. Daly" has presented the requisite
data for estimating the area, ice-covered during the Pleistocene,
which is now freed from its glacial burden. If the ice averaged
5,000 feet in thickness over this area, the sea-level would have been
lowered about 235 feet by abstraction of water. In comparing
this figure with the 136 feet of elevation at the border of the assumed
Labradorian ice sheet it should be remembered that a considerable
body of Pleistocene ice existed at distances of more than 4'5° from
its border and would have exerted an attraction in the opposite
direction.
It seems safe to conclude that at no time or place was the
gravitative attraction of a Pleistocene ice cap powerful enough to
elevate sea-water at its border sufficiently to overcome the lowering
of sea-level due to withdrawal of water from the sea. This means
that so far as these two factors are concerned the movement of the
strand line was everywhere negative during the time of advancing
ice and positive during the waning stages.? Near the ice borders
the movement was in each case less in amount than in lower lati-
tudes, but its direction was the same. Disregarding secular adjust-
ment within the earth, the retreat of the Labradorian ice front from
Covey Hill northward must have resulted in a progressively more
extensive submergence of the St. Lawrence-Ontario Valley.
For present purposes it is unnecessary to make any inquiry
into the causes of the differential uplift of northern lands which is
known to have occurred since retreat of the Labradorian ice sheet
began. Itis, however, essential to know the time relations between
the withdrawal of the ice and the readjustments which so convin-
cingly appear to be of an isostatic nature. Did the land mass
respond so quickly that secular movement entirely compensated
the rising sea-level and thus maintained a stationary, or even a
retreating, strand-line? Or did crustal deformation lag behind
unloading of the ice burden sufficiently to permit an upward move-
ment of the sea-level to precede the upward movement of the land ?
t™R. A. Daly, ‘‘The Glacial-Control Theory of Coral Reefs,” Proc. Amer. Acad.
Arts and Sci., L (1915), 172.
2 Cf. C. Schuchert, ‘“‘The Problem of Continental Fracturing and Diastrophism
in Oceanica,” Amer. Jour. Sci. (4), XLIII (1916), 92, 93.
THE CHAMPLAIN SEA IN THE LAKE ONTARIO BASIN 553
The philosophy of diastrophism is not sharply enough defined
to permit an a priori answer to this question. On the one hand
are many data, such as those obtained by Michelsont in his study
of tides, which indicate quick response to external strain; on the
other hand are just as many, if not more, facts, such as those con-
. cerning periodicity of orogenic episodes, which indicate the ability
of the earth to delay adjustment for some time after strains begin to
accumulate.
Field evidence in the Great Lakes region strongly suggests con-
siderable lag of continental uplift behind ice removal. Post-
glacial marine fossils in the Hudson’s Bay drainage basin occur at
elevations at least as great as 450 feet? and indicate uplift of that
amount since the Labradorian ice sheet shrank to a diameter of
less than two or three hundred miles. The highest shore line of the
Champlain Sea rises toward the north at an average gradient of
more than 23 feet per mile and is essentially parallel to the Iroquois
strand—sufficient proof in itself that the uplift of the St. Lawrence
region, in greater part at least, lagged considerably after removal of
the ice load.
The facts concerning Quaternary diastrophism in the region
covered by the Labradorian ice sheét seem to be in complete agree-
ment with the conclusion reached from field evidence in Napanee
Valley, namely, that the late Wisconsin and Recent uplift did not
cause the marine strand to retreat until after an appreciable interval
of sea-advance had resulted from the melting of the ice.
1A. A. Michelson, ‘‘ Preliminary Results of Measurements of the Rigidity of the
Earth,” Jour. Geol., XXII (1914), 97-130.
2 A. P. Coleman, ‘‘Lake Ojibway; Last of the Great Glacial Lakes,” Ontario Bur.
Mines, Ann. Rep., XVIII (909), 284-93.
3 In determining the amount of deformation affecting the shore lines of an inclosed
body of water, such as Lake Iroquois, during its existence, the distortion of its surface
by gravitative attraction should by no means be disregarded. Retreat of the ice
front will, in effect, carry the inclined plane of the water surface northward and at the
same time cause it to approach more closely a horizontal position. The result will be
splitting of beaches north of the lake’s outlet, and drowning of the older shore features
south of the outlet. This apparent, though not real, warping of the basin was in many
cases of sufficient amount to be of quantitative importance. One reason why the
Nipissing beach, although tilted, departs but little from a true plane, while the Algon-
quin and earlier beaches are warped as well as tilted, is that Lake Nipissing was not
a marginal glacial lake, as were its predecessors.
554 KIRTLEY F. MATHER
SUMMARY
Withdrawal from the Thousand Island region of the Labradorian
ice barrier responsible for the existence of Lake Iroquois was fol-
lowed by accumulation of fluvio-glacial gravels in Napanee Valley.
The pro-glacial stream descended to a locality which is now less
than 325 feet above sea-level before it debouched into Gilbert
Gulf, an arm of the Champlain Sea. At the same time Algonquin
River carried the overflow from Lake Algonquin down Trent
Valley past Trenton, Ontario, to an outlet which is now beneath the
waters of Lake Ontario.
The melting-back of the ice front from Covey Hill toward the
Height of Land was contemporaneous with a positive movement of
the strand-line which carried marine waters toward the head of the
Ontario basin and drowned the Napanee and Trent valleys. The
positive movement of the strand-line was followed by a negative
movement, which began approximately at the time of complete
disappearance of Labradorian ice and has continued nearly or
quite to the present time.
Disregarding crustal movement, the waning of Pleistocene ice
caps would result in a world-wide transgression of the sea as its
volume was increased by the return of water temporarily abstracted
to form the ice masses. In high latitudes the amount of movement
of sea-level would be much less than in low, because of gravitative
attraction of the ice, but everywhere the direction of movement
would be the same.
Secular adjustment following removal of ice load was delayed
in the Ontario-St. Lawrence region long enough to permit a stage
of sea-advance before upward tilting overcame the effect of ice-
melting and the stage of sea-retreat was reached.
THE RELATIONSHIPS OF THE FOSSIL BIRD
PALAEOCHENOIDES MIOCEANUS
ALEXANDER WETMORE
U. S. Biological Survey, Washington, D.C.
In a recent number of the Geological Magazine’ Dr. R. W. Shu-
feldt described and figured the fossilized distal end of the right
femur of a bird from South Carolina, proposing for it the name
Palaeochenoides mioceanus. This Dr. Shufeldt considered as
representing a large anserine form. ‘Through the courtesy of Dr.
O. P. Hay, I have had opportunity of comparing this type, and after
careful study am forced to disagree with Dr. Shufeldt as to the
affinities of the species represented. After careful. comparison
with many specimens I am convinced that the fragment does not
come from an anserine bird, but that it represents a large stega-
nopod, related (though not closely) to our modern brown pelicans.
When the specimen was first examined, the popliteal area of the
bone was obscured by matrix that covered and obliterated contours
and slight depressions. This was carefully removed, and these
characters made fully visible (Fig. 1). Though at first glance there
are certain resemblances to the swans, these characters are found to
be superficial and to lose their value upon careful study. The
anseriform species available that show certain resemblances to Pa-
laeochenoides are the following: Olor buccinator, O. americanus, O.
cygnus, Branta canadensis, Chen caerulescens ,and Dendrocygna autum-
nalis. A considerable number of other species have been examined,
but have been found to resemble the above closely or to be so differ-
ent as not to be pertinent in the present case. In the Steganopodes
the following have been utilized: Phaéthon aethereus, Fregata magnifi-
-cens, Sula leucogastra, S. bassana, S. serrator, Pelecanus fuscus,
and P. onocrotalus. The cormorants and darters are highly
1916, pp. 343-47 (Pl. XV).
555
556 ALEXANDER WETMORE
specialized, so that, though several species of Phalacrocorax, Nan-
nopterum, and two species of Anhinga were examined, they were
not directly comparable. In the following table I have drawn up
in parallel columns the salient differences in the distal end of the
Fic. 1.—Lower surface of distal portion of femur (type) of Palacochendides
mioceanus.
femur in the species enumerated in the two groups. An asterisk
indicates which group Palaeochendides resembles in the characters
designated.
Anseres Steganopodes
t. Intercondylar notch deep,narrow.* 1. Intercondylar notch more shallow,
2. Femur non-pneumatic. broader.
3. Condyles well elevated on dorsal 2. Femur pneumatic.*
surface, rising abruptly from shaft. 3. Condyles little elevated on dorsal
4. Ventral surface of femur behind surface, merging in a long gradual
condyles more rounded, lateral slope into shaft.*
margins strongly rounded. 4. Ventral surface of femur behind
5. Tuberosity above fibular facet of condyles flattened, lateral margins
outer condyle extending at an angular or very slightly rounded.*
angle across outer third of shaft. 5. Tuberosity above fibular facet of
outer condyle lateral, following
line of shaft.*
It is seen that in four of these major differences Palaeochendides
agrees with the Steganopodes, while in only one does it approach
the Anseres. Other differences of less constant value are present
between the two groups. In most of the anserine birds the con-
PALAEOCHENOIDES MIOCEANUS 557
dyles project farther on the ventral surface, and the distal end of
the femur is greatly expanded at the condyles, the shaft being
slender in comparison. In the Steganopodes (save in the Phalacro-
coracidae) the condyles are less produced ventrally and there is a
gradual broadening of the shaft until it merges gradually into the
condyles. In the Anseriformes, in general, the lateral diameter
of the shaft of the femur where the expansion for condylar support
ceases is less than one-half the greatest lateral width through the
condyles (the measurement of the shaft in this case not being abso-
lutely its smallest diameter, but usually the breadth at a point
one-third of the length of the femur from its distal extremity).
In the Sulidae, the brown pelicans, and the snake-birds this diam-
eter is more than one-half of the condylar width. Some cormorants
have it greater (P. albiventris), and some less. In all of these points
Palaeochenoides resembles the totipalmate birds, and it is referred
without question to the Steganopodes. The distal end of the
femur representing Palaeochendides, while similar to that of our
present-day pelicans, differs in having a posterior pneumatic
foramen, the popliteal space divided by a rounded elongate ridge
(extending at an angle posteriorly from the pneumatic fossa),
the outer condyle broader and stronger, and the intercondylar
channel deeper and more narrow, with no depression evident on the
dorsal face of the shaft immediately posterior to the origin of the
condylar ridges. Should more of the skeleton become known, it
may eventually be placed in a separate family. If we may venture
to base theory upon this one fragment, Palaeochendides was a
pelican-like bird somewhat larger than Pelecanus erythrorhynchos
or P. onocrotalus, as the portion of the femur representing it seems
to indicate that the bone in its entirety was somewhat larger and
heavier than the femur in these two species. In its appearance
this bone seems, too, to show certain resemblances to the Sulidae
and remotely to the Anhingidae and the Phalacrocoracidae.
Hence, while Palaeochendides will stand as a milepost in the line of
descent of the pelicans, it brings down to us suggestions of general-
ized development indicating ancient relationships of pelicans to
gannets and more remotely to the cormorant-anhinga branch of
the totipalmates.
A STUDY OF THE FAUNAS OF THE RESIDUAL MISSIS-
SIPPIAN OF PHELPS COUNTY (CENTRAL
OZARK REGION), MISSOURI
JOSIAH BRIDGE
University of Chicago
In the reports of the early writers on the geology of the Ozark
region in Missouri mention is made of scattered patches of bowlders
containing fossils of ‘“‘Chemung age” occurring in counties where
no rocks of this age are in place or where they are but sparingly
represented.
The investigations of Shumard, Meek, and Broadhead for the
Missouri Bureau of Geology and Mines, between 1855 and 1871,
published in 1873, mention such bowlders from Maries, Miller,
Morgan, Phelps, Wright, and Ozark counties. This list of local-
ities is doubtless incomplete, since much of the region was not
examined by these workers. Meek assigned these bowlders to the
Lower Carboniferous, but Shumard and Broadhead employ the
older term Chemung. Shortly before the publication of the report,
the so-called ““Chemung Group”’ of the older Missouri geologists
was shown to be Lower Carboniferous, and in a footnote in Shum-
ard’s report on Ozark County,” he states that the term Chemung
as employed in these reports is to be understood to mean Lower
Carboniferous.
Broadhead again mentions these deposits in his report for
1873-74, as follows: ‘‘Fragmentary outliers of this group [Chou-
teau] are occasionally found capping the ridges near the Arkansas
line.’’3
™ Missouri Bur. of Geol. and Mines, Reports of the Geological Survey of the State
of Missouri, 1855-71 (1873). See reports of the various counties named above.
2 Ibid., p. 190.
3G. C. Broadhead in Missouri Bur. of Geol. and Mines, Reports of the Geological
Survey of the State of Missouri, 1873-74 (1874), p. 27.
558
MISSISSIPPIAN OF PHELPS COUNTY, MISSOURI ~ 559
In 1898 Broadhead, writing on the Ozark uplift and the growth
of the Missouri Paleozoic,’ states that ‘‘beds of Chouteau lime-
stone’’ have been found in Wright and Maries counties, and that
the Burlington limestone has been recognized in Morgan, Benton,
and Hickory counties, from which he concludes that at least the
western half of the Ozark uplift was submerged during the Lower
Carboniferous period.
The work of the present Missouri Bureau of Geology and Mines
has given some additional information concerning the distribution
of these deposits. Marbut,? Ball and Smith,3 Van Horn and Buck-
ley,4 and Lee,’ working in different areas in the northwestern part
of the Ozark region, have mentioned these deposits, but up to the
present time no careful study has been made either of the distribu-
tion of this residual material or of its faunal content. In the litera-
ture it is commonly referred to as Chouteau or Burlington. In
some localities these deposits are termed “‘residual Boone Cherts.”
The purpose of the present paper is to present the results of a
careful study of the faunas contained in a number of these bowlders.
The bowlders which have furnished the material for this investiga-
tion were all collected within an area of about forty square miles
situated in the extreme northeastern part of the Rolla quadrangle
and the adjacent portion of Phelps County.
This area is a small plateau the summit of which is capped by
resistant formations. The physiography and general geology
of the region have been well described by Lee® and need not be
rediscussed here. The entire plateau is underlain by the Jefferson
City dolomite of Canadian age, and lying unconformably upon
1 G. C. Broadhead, ‘‘The Ozark Uplift and the Growth of the Missouri Paleozoic,”
Missouri Bur. of Geol. and Mines, Vol. XII (1898), Areal Geology, p. 308.
2 C. F. Marbut, ‘‘ Geology of Morgan County,” Missouri Bur. of Geol. and Mines,
Vol. VII (1907), 2d ser., pp. 49-51.
3S. H. Ball and A. F. Smith, ‘‘Geology of Miller County,” Missouri Bur. Geol.
and Mines, Vol. I (1903), 2d ser., pp. 82-89.
4F. H. Van Horn and E. R. Buckley, ‘‘Geology of Moniteau County,” Missouri
Bur. of Geol. and Mines, Vol. III (1905), 2d ser., pp. 44-58.
5 Wallace Lee, ‘“‘Geology of the Rolla Quadrangle,’ Missouri Bur. of Geol. and
Mines, Vol. XII (1913), 2d ser., pp. 41-43.
6 [bid., pp. 52-58, 13-40.
-
560 JOSIAH BRIDGE
this are patches of non-fossiliferous sandstone and fire clays, which
on lithologic grounds are classed as basal Pennsylvanian. The
area covered by these Pennsylvanian deposits is not definitely
known, but recent work has shown that its extent is much greater
than has heretofore been supposed. It probably caps the divide
almost continuously from Rolla to Cuba and for a considerable
distance northeast of Cuba. Around the edges of this plateau
where the Pennsylvanian is wanting, but where the Jefferson City
formation is still in place, there are found small areas which are
covered with Mississippian bowlders. ‘They rarely cover more
than a few acres and may or may not be associated with similar
deposits of Pennsylvanian age. They have an average vertical
distribution of about thirty-five feet and a maximum of eighty feet.
The total relief of the Rolla area is about five hundred feet, but
the bowlder patches are confined to the upper two hundred feet.
They occur at lower levels in the northern part of the area, owing
to the slight dip of the underlying formations.
The bowlders almost invariably occur on the hillsides or in
the heads of small ravines. A few patches have been found upon
the hilltops, but in every case the crest of that particular hill was
below the general summit-level. In the places where one of
the bowlder areas is associated with a small outlier of Pennsyl-
vanian it may completely encircle it, but in most cases forms a
fringe on one side only, which would seem to indicate that these
bowlder areas were localized in some manner before the deposition
of the Pennsylvanian.
The exact nature of the Mississippian-Pennsylvanian uncon-
formity is still doubtful. Outliers of Mississippian rocks occur in
some parts of the Ozark region, but up to the present none have
been found in this area. Small isolated patches of stratified Missis-
sippian may be buried beneath the Pennsylvanian, but at the present
time no such occurrences have been reported. At one locality
fragments of the Mississippian have been found in the basal con-
glomerate of the Pennsylvanian,’ but such association is not com-
mon. It may be that these patches of bowlders are the remnants
of small outliers which have been reduced to this stage since the
1 Wallace Lee, op. cit., p. 44.
MISSISSIPPIAN OF PHELPS COUNTY, MISSOURI 561
erosion of the Pennsylvanian, but there is also the possibility that
these deposits were in the bowlder form before the invasion of the
Pennsylvanian seas. In either case the character of the bowlders
seems to indicate that they have not been moved far from their
place of origin, if they have been moved at all.
Lithologically the bowlders fall into a number of well-marked
groups. The more abundant ones are of course, dense, quartzitic
sandstone, which in its original state was probably a calcareous
sandstone. In their present condition the calcium carbonate
content has been completely removed by leaching, and thin sec-
tions show them to be made up entirely of clear quartz grains
exhibiting little evidence of secondary growth. The interior of
these bowlders is semi-translucent, bluish white in color, and the
rock is quite hard and dense, but the more weathered, external
portions are extremely porous and deeply stained with iron oxide.
They are abundantly fossiliferous, but the fossils are scattered
irregularly through the rock with no arrangement that would
suggest bedding. Associated with the quartzitic bowlders are
masses of soft, friable, fine-grained white sandstone, which are
somewhat leached and in which the more weathered phases are
deeply stained with iron oxide. Otherwise the rock is but little
altered. The fossils of these sandstone masses are arranged in
parallel bands, which probably correspond to the bedding planes
of the formation.
Bowlders of chert are also abundant. Some are practically
, unweathered, others are so completely leached that they powder
under the hammer. For the most part they are white or pale
bluish white in color, and, like the sandstone bowlders, they are
stained with iron oxide in proportion to the amount of weathering
which they have undergone. At one locality a single chert bowlder
has been found which is of a pale pink color, the color appearing
to be original.
Lee’ has mentioned that bowlders of siliceous odlite of Mississip-
plan age occur at one locality associated with bowlders of the
quartzitic type. A careful examination of such odlite bowlders
from this and other localities where they are associated with residual
1 [bid., p. 42.
562 JOSIAH BRIDGE
Mississippian deposits has failed to disclose any determinable
Mississippian fossils. Some of the odlite is conglomeratic, con-
taining small angular chert pebbles, and much of it contains small
cavities, some rounded, others angular, which simulate the impres-
sion of poorly preserved fossils. Lee has suggested that this odlite
is the possible equivalent of the Short Creek odlite of southwestern
Missouri. The writer has not seen Lee’s collections from the fossilif-
erous oGlite, and his own observations have not confirmed Lee’s.
Bowlders of siliceous odlite are abundant in the residual material
over a large part of this area. There are beds of it in the Jefferson
City formation, and some odlite bowlders have been found which
contain Canadian fossils. On the other hand, many of the odlite
bowlders are weathered and stained with iron oxide in much the
same manner as are the Mississippian bowlders, while others do
not seem to have been weathered in this manner.
The association of the different types of bowlders seems to
follow no general rule. In some localities they consist exclusively
of sandstones, while in others chert predominates, but in nearly
every case examples of all the easily recognized lithologic types are
to be found together.
The fossils are in most cases preserved as molds of the exterior
and as casts of the interior. In rare instances the shell itself,
together with the internal structures, has been completely silicified.
In many examples the molds and internal casts preserve the mark-
ings of the original with great fidelity, and excellent squeezes may
be obtained from them. Some difficulty has been experienced in -
correlating the mold with the internal cast, for, strangely enough,
when one surface is well preserved, the other often is not.
The material used in the preparation of this paper has come
from three sources: (1) a collection belonging to the Department
of Geology of the Missouri School of Mines and Metallurgy, lent
by Professor Cox and Professor Dake; (2) some collections be-
longing to the Missouri Bureau of Geology and Mines, lent by
Mr. Buehler; and (3) a number of collections made by the writer
and his friends. ‘These were made while the writer was connected
with the Missouri School of Mines and are the property of that
institution.
MISSISSIPPIAN OF PHELPS COUNTY, MISSOURI 563
At the outset it was realized that in order to get decisive
results the faunas of each individual bowlder must be kept together
so that the natural association of species might be determined.
This has been done, and a record of the localities from which the
individual bowlders have come has also been kept. The original
collection of the School of Mines was made without regard to the
first factor and with but slight regard to the second, so that, while
it contains many fine specimens which have been of great value in
making comparisons, it has been useless for exact stratigraphic
purposes. The collections of the Missouri Survey and those made
by the writer were made with both of the above-mentioned points
in view, and all of the conclusions drawn from this study are based
upon the faunas which they have yielded. Numbers have been
assigned to the several localities from which collections have been
made, and these, together with the location, are given in the accom-
panying list.
REGISTER OF LOCALITIES, PHELPS County, MISSOURI
38 N., R: 8 W., N.E. 7, S.E. 2, Sec. 36.
38 N., R.8 W., NE. 4, NE. 3, Sec. 36.
38 N., R. 8 W., S.W.z, S.E: @, Sec. 25:
37 N., R. 8 W., N.W. 4, S.W. 2, Sec. 26.
37 N.; R. 8 W., Center line S. 3, Sec. 25.
38 N., R. 8 W., S.E. 4, S.W. 4; Sec. 27.
38 N., R. 8 W., W. 3, S.W. 3, Sec. 27.
37 N.,-R. 8 W., S.W. 4, S.W. 4, Sec. 24 and adjoining corners in
Secs. 24 and 25.
9: ©. 38 N.,R. 8 W., S.W. 4, SE: 4, Sec. 35.
10. T. 38 N., R. 8 W., N.W. corner N.W. i, N.E. d, Sec. 35.
1z. T. 38 N., R. 8 W., N.W. 4, N.W. 4, Sec. 36.
12. T. 38 N., R. 8 W., Center S.E. 4, Sec. 26.
SOON eS Es
MARIES COUNTY, MISSOURI
is) deyAO NE Rao Wis SC. 7.
Lee’s map shows the location of most of the localities listed
above.
The oldest Mississippian fauna which this area has yielded is
contained in the bowlder of pink chert, mentioned on page 561.
The bowlder is case-hardened and stained with iron oxide to the
564 JOSIAH BRIDGE
depth of about one-eighth of an inch. It is somewhat leached, but
otherwise is scarcely weathered. The following fauna was obtained:
FAUNA OF BOWLDER 28, LOCALITY 7
Coelenterata
Zaphrentidae
Gen. ? sp.?
Cyathaxonidae
Cyathaxontia sp.
Echinodermata
Platycrinus sp. (stem fragments)
Unidentified stem and other fragments
Molluscoidea
Fenestella burlingtonensis Ulr. ?
Rhombopora sp.
*Productus sampsoni Weller ?
Rhipidomella diminutiva Rowley
Camarotoechia tuta (Miller)
*Dielasma fernglenensis Weller
Terebratuloid shell gen. ? sp. ? (3 forms)
*Cyrtina burlingtonensis Rowley
Spirifer sp. (small form, probably young shell)
*Spiriferina subtexta White
*Brachythyris fernglenensis Weller ?
Ambocoelia levicula Rowley
Ambocoelia sp.
Reticularia cooperensis (Swallow)
*Ptychospira sexplicata (W. &. W.)
*A thyris lamellosa (Leveille)
*Cliothyridina glenparkensis Weller ?
A comparison of this fauna with the Fern Glen fauna described
by Wellert shows that the two have very close relationships.
While the fauna of this bowlder is not as large as that of the
Fern Glen, the majority of its determined forms are identical with,
or closely related to, Fern Glen species. In the faunal list given
above the species occurring in the Fern Glen fauna have been
designated by an asterisk. Of the ten certainly identified brachi-
opods, six are recorded in the Fern Glen fauna, and three others,
doubtfully identified, are also represented in the Fern Glen. The
«Stuart Weller, ‘Kinderhook Faunal Studies, V, The Fauna of the Fern Glen
Formation,” Bul. Geol. Soc. Amer., XX (1909), 265-332.
MISSISSIPPIAN OF PHELPS COUNTY, MISSOURI 565
genus Cyathaxonia is abundantly represented in the Fern Glen
fauna, and Platycrinus and Rhipidomella also occur. Camaro-
toechia tuta and Ambocoelia levicula are both found in the Lower
Burlington limestone, while Reticularia cooperensis appears to be
a survival from the late Kinderhook. ‘The relationships of this
fauna are so definite that there is no hesitation in calling it the
equivalent of the Fern Glen fauna.
A somewhat younger fauna has been obtained from a number of
quartzitic bowlders, and the faunas of the four best bowlders con-
taining this fauna have been combined to form the list given in
Table I.
A comparison of the four faunas listed (p. 566) shows that
they are closely related. Only one species is common to all four
faunas, but three species are common to faunas Nos. 3 and 5, seven
to faunas Nos. 31 and 5, five to faunas Nos. 17 and 5, etc. The
bowlders in which these faunas were contained were all small, and
this would partially account for some of the diversity.
The affinities of these faunas are with the Burlington rather than
with the Kinderhook. Many of the species are common to the
Upper Kinderhook and the Lower Burlington, and recent work
has suggested that some beds which have commonly been referred
to the upper part of the Kinderhook are more nearly related to the
base of the Burlington. This fauna is not the equivalent of the
Fern Glen fauna just described; neither is it the exact equivalent
of the fauna of the Burlington White Chert from Louisiana, Mis-
sourl, but appears to be intermediate between the two. Very
few faunas have been described from this horizon, and, until more
definite work is done on this portion of the Mississippian, a more
exact correlation cannot be made.
The faunas of the eight bowlders listed in Table II have much
in common, and undoubtedly are from a common horizon.
Of the 72 forms listed in Table II, 50 are found in the Burling-
ton of various localities, and at least 11 more occur in the Fern Glen
and in beds of equivalent age. A few forms hitherto reported
only from the Upper Kinderhook and Chouteau—Chonophyllum
sedaliense, Reticularia cooperensis, Productus blairi, and Hustedia
circularis—are also recorded. Some of these are definitely identified ;
566 JOSIAH BRIDGE
the presence of others is doubtful, but their presence does not
affect the correlation of the faunas. Some of the bryozoans listed
TABLE I
Locality Nope eens iti Meier tena! et cae
BowldernNoweci verre iets stores eaetorveeste rate eterat
Coelenterata
Zaphrentis calceolaW.& W........
Echinodermata
Cryptoblasius melo OVE Ss. cin 5 os oteiee ok noes
BIGUYCEINUST ZASD in Ses eis ee eysyere ee tal tue, axon tints
Undetscrinoidibase ime te eee. eee
Molluscoidea
Henesicllaxchy hvexveua Weer Mone lh ee sc
Roly POG 'SPi esi uaere one ene Mec t2t el eo] levees she iat
SehellwtenellaiSpinceyaicwtes pesos sc:
Chonetes multicosta Winchell........|.........
Ghonetes spaundesc iia csnce lie se ena:
Productus sampsont Weller. ........|.........
Productus fernglenensis Weller.......
Productus burlingtonensis Hall......}.........
PRR OGUGCLUSASD EN AER ie ORI Care ee Lahaye Lease ee Pod [eee cues
Rhipidomella diminutiva Rowley.....
Schizophoria chouteauensis Weller... .
Schizophoria swallovi (Hall).........
tal
Schizophoria subelliptica (W. & W.)..|.........
SGHLZO PHOFLONS Pde ee sete creek ee uefa TON Merge el Rr As Seen aa
COMmGrotoechraispu rnin dena ler cues ge fa OMS or ano leat Area ae
IRR HCHO PORGISD | ius ve oh cs ahe SUH SIRE ae ne IAA ANE La
Crancenanolobosa Weller een | ne Mdm Neti sa
Dielasma vosceolensis) Weller te ey. ee
PIT CLUE SD eR mene ie pereyeve de lence tens testa o | Metal abe
Spirtifer osagensis Swallow..........
Spirifer vernonensis Swallow........
Spirifer latior Swallow.............
Spirifer louisianensis Rowley.......
Spirtfer legrandensis Weller.........
Spirifer platynotus Weller..........
SWE AWG? LUO ES Waar sine aah ellbannes ooulloodcoauee
PY EAA Oy ANON erected sear sie MRE seen hy cilleiae pees SOB MOCO GEE OS
SPUD CVASD Gr pane NUS einer oe Lem el Ni cesctelrcnalle dena cetera
PUTTER SD iy gaieeetses oerausee choca cme nemenliecatst eheuS Ne eueteatnarers
(BY ACHYLEN TIS NCES DR nisin te eles
Syringothyris Sp........ Rebeka ahaa
Ambocoelia levicula Rowle
Nucleospira obesa Rowley..........
Cliothyridina tenutlineata (Rowley) ..
Mollusca
Conocardium sp. NOV... 4... ..2.
Euomphalus latus Hall.............
Platyceras obliquus (Keyes).........
Orihonschiaspy cue ee eae
ELOGET ES, Zu SP ease a elapse clays repens te eeuel| be suai erat eve seen aayeaece
Arthropoda
Phillipsia meramecensis: snumard ini. oo). ae alle elon
Io
wee puiie jateKelic
eis 0) 8/\s/ ele elre'
Cyr) On ORCC Ong
ORONO O10
MISSISSIPPIAN OF PHELPS COUNTY, MISSOURI
TABLE II
567
EO CALI GYAN MN eine en sete neyedave tay shantancl oe iencnats ais
Bowlder No
Coelenterata
Zaphrentis calceola W. & W...
Zaphrentis sp. undet
Zaphrentis sp. undet
Zaphrentis sp. undet
Triplo pln centralis
(E. & H.)
Tri ‘plophyllum sp
Amplexus fragilis White &
ot. John
Chonophyllum sedaliense White
Echinodermata
Cryptoblastus melo O. & S.....
Batocrinus ? tuberculatus ?
W.&S
Aorocrinus ? ? sp
Actinocrinidae gen.? sp.?.....
Platycrinus sculptus Hall
Platycrinus sp. (several)
Crinoid fragments, plates,
stems
Molluscoidea
Fenestella filistriata Ulr
Fenestella multispinosa Ulv.. . .
Fenestella compressa vat.
elongata Cumings
Fenestella burlingtonensis Ulr
Fenestella ? sp
Fenestella? sp
Fenestella? sp
Hemiirypa? sp
Fenestelloid gen., sp
Rhombopora sp
Gly ptopora sp
Leptaena analoga (Phillips). . .
Schellwienella alternata Weller.
Schellwienella inflata W. &. W.
Schellwienella burlingtonensis
Weller
Schellwienella or Streplorhyn-
chus sp
Streptorhynchus sp
Chonetes glenparkensis Weller
Chonetes multicosta Winchell. .
Chonetes illinoisensis Worthen
. Chonetes logani N. & P
Chonetes sp. nov
Chonetes sp
Chonetes Spyies seeders
Productella concentrica (Hall). .
Productella millespinosa Girty
Productus ovatus Hall
568 JOSIAH BRIDGE
TABLE T—Continued
Locality: Noi eo ee RU Ears I 4 5 Io 10 Io I 12
Bowlder Noss eis aia aoa toe Meas MGS | tor 100 4 2 27 MGS | MGS
2 I 3
Molluscoidea—Continued
Productus sampsont Wellerie. ix il clei) easel ee al eaet ete Xe x
‘Producius: sedalzenstsiwelleris ais cisin |feeveue eee elles eee eee cf.
Productus blairi Miller........ CE SO OSE Ras LAG ted Ut era adel ome OR A aN
Productus burlingtonensis Hall|..... bE Reefer ene ue yer| PS uct x
Productus fernglenesis Weller..|..... Folin Repent eit boop ie tee Duell erenete
EATOGUGLUS TSP Ute e UeN a a ea ree werlies svetsay|icirescttat Fas dds ghe Ae oceesl | ere ueeed| ever tours ogee x
Rhipidomella burlingtonensis a &
CELA er Vaart tase fea let p40 [ass ed retary Pe Oh Aba 20 Nis
Rhipidomella diminutiva x&
Rowleyegienc sae veey sete x bq I a hee ATTN soci ie WS ca aaa cf.
IRI pid omella thcencer AWW Ce) asl 005s oiee) 03 aie eee elec rcues eel mane x
Rhipidomella sp pene er ellie Sea a cael eave allie veeearcll Mirco tcl ated | enon
Schizophoria chouteauensis
Wellenin sire tenet pier 5 CEP ee earaca| ie Me ees] ee Nramelf Auer a nel ene eae | ear x
Schizophoria swallovi (Hall)...|..... bali | eerie HEE Puebel Ure se eee kes Seria Hnacpriee DA OL
Schizophoria poststriatula
Weller ieee ar eee losrae Ochi eeu has| audeoare | alec saat arayee con aeeenaeu | Bee x
Camaro phoria bisinuata
(Rowley sya estintia ee tsn at teri lencler gl ly i aMt i aciidoaeles [or Mecca ae b:don Weis apatig eh es
Camarotoechia tuta (Miller)...|..... Dine air ei eee Yaa Lem aE ane! x
(COTTAGE OE Se ona notes Slee o walle AB nsoee allaomolloeodulleosdc Xl ea eas
Rhynchopora persinuata
(Wil Ghiell) Sos pau ska ticle se) aeecail Nn ciaaa i rayep sll era Gall te cde glare | oe Wee a | Mere x
Centronelloidea rowleyi
@Worthen) Aan ae boli CanapernlVeneaty ra Le ae ie Rte eae
CranaenatelobosanWiellerays sai cs Hn |eeiaen | les torent ale (anata aves Kal isaeae
DiclasmaGhouteanensiss Weller|e ix) Gisela slice fella ee line ee el lees cell oan etl eae
Dielasma osceolensis Weller. ..|..... pCa ease Bet Papel araee] Ps cereal eee el ey aan La
Dirclasm aspen ate KOI coh sett] steele reat | eae | a ete | eee
Spiriferina subtexta White....| x XNA. oe fara Ree al eae x i aleueerets
PIG CNLLGN SOLIALYOSURIS INV DALE risieta allie cvei-tal| eevetaaleelon at. test | eee |e X
PO HEA AGES OMIR He araies Gialeietin Gia Roe eo lal ha eaeNel ee eI Et Gieneaie x pil Wael Vetere
Delthyris novamexicana (Miller)|..... AA Sratey Nena Si sa aan a irra cuaratanel eae
Spirifer louisianensis Rowley..| x x bail Phere ya Nena ual viet x x
Spirifer vernonensis Swallow..| ? Kea Mla ieee: [bas cue lt sea tee | esac ? x
DYLAN GOT ALS AROS NA Saleoe) lmoaaeleosie alana s| ea kaulleasdolloeacullgoonc
Spirifer platynotus Weller ....| ? Folin ennie artrt [eete eA Siam eve ee eal Ku onosrs ?
Spirifer biplicoides Weller ....|..... SRM MISA areca Svea eens al rete | ae eg x
Opirifereregera Weller’... alae. Dele Fal Sie Ser] a Et Ila beeen ips Neve geri
Spirifer osagensis Swallow....|..... 5 ded ramen hocerssl rae es a Sate x x
SPUNK enONDES DINER CAME ey ie oiel veya Naneunent | evens Ba ava ve ets x x
SY EG AA AALS ROME Goel ds sullbeosoldibeealocedieonee lode: Perel eileeos
SPL CTArOWLEVE, NV Elletia: Ales al uralonn |W ceases |e case eceille Ciena aria [raat eee ee x
SAAD cle oitie Subd ee CEG eae pais rears aver genta tees eae ell betel ee lc oes bc
SPILT ROPES Devin sieve ada Winery, Aslan bata eRe al EAT al Livin el es etaecII Maa ell wate a
POY EAI ARS) Dinettes eee Bes a cree ene her en Rieti Vel nero esReeeeteal fa oko ice lacletenes teaiiehena Po crovate 3
SYA MG? Soe (Groene val) ae eee oleae ul llooaclb edule alls aoe: a x
SPIRAL ER USD gue e One aN oun a tars 2s letra ete Lie cr cee [ean BSI Ra eet Ket eg
Brachythyris chouteauensis
Wiellerii Orrin aeanrolay a D.C Tce aati Shyer] Mana annie ae Paced lean ee x x
SVELMZOLAY ISIS Nero tiecc eens, ic | eye eee | eee efea | ayo (pee eee gaara ise | RSe x
MISSISSIPPIAN OF PHELPS COUNTY, MISSOURI 569
TABLE II—Concluded
HOCALILVRINO epi ce evstaloie tare alesteehveacseoe eke I 4 5 Io Be) Io I 12
BowldertNossaiic ee carats mietcessusinecere a ieake MGS | tor 100 4 2 27 MGS | MGS
Mulloscoidea—Continued
Pseudosyrinx missouriensis
AWiell exe aarey eeu sea tet MO AON eS MAEM Sc Gut PANN ese
Ambocoelia levicula Rowley...| x |..... bo ata Ieee a x Fee eeataistal ene
Reticularia cooperensis (Swal-
LEUMELNIG CSP ered cece er il PRAM see arto [pe [AR Uae re ae ba ea ee Rain ia Gara
iHustediacircularis: (Maller. oe oes eee nile ects = pani SV aie aaa
Nucleospira obesa Rowley..... beige 2A MO socal me ve FU eR
Rowleyella fabulites (Rowley). .|.....|.....).....).. EVAN ai, Wanner neh
Athyris lamellosa (Leveille)...|.....).....|-. STENT Ne a Fea EN 9-11 ee a
Cliothyridina tenutlineata
(Rowley)...... FLMe uaa x SHOP siemnensesy [pe em ec Ne ULNA x x
Composita sp is eee ooaiey anes Oy PT OU) Tce a ee Re
Mollusca
Conocardium sp. nov........|...-- x DS Ke sats eer Pg Heep soar tel a a
CY PHCOLCINTE SD Wayne iat ye lieth el orereay all Nae tee stan (Beau ill a ade ag x x
Laevidentalium sp........... Bp Nal aa IS tic He eer | CUE AS | Vea ee tea NSM Fe
ILA SIS B03 Ouida Sac eacass bin LS re oS] (ea ee eg Pentti Peseta Pe ESN A AP
Pleurotomaria sedaliensis Mil-
Platyschisma? depressa......|..... SE A Sa Ser EMU a [Se ele payer acs paul fsa
Euomphalus latus Hall.......)..... x Dial Fes ULE bec hes ah IB ease x x
Strophostylus bivolve W. &.W..| xX |.....|...--).---- KRU VA MeN als Neg x
OniLOWNGHY GES set oh ehh lls eae | aetna ge (a ATS TEE le ne TEs x
Oxthonychiaisp ys yes ve) Se bead Umea De ai ener real e Sass x
Platyceras nasutus Miller.....|..... Fa escltahen NEU tpl Peale es a Mees Ue ey
JEON RGHOS HUTA NIG koe WEE atl leis eg al apna aoe ool dalla on ae Alia seo es
WAL OLY GEX GSH ODLIQUUS NNEV. CS ars cislec roan al cesta el evan enel| einncbe eiliere lesa [yeaa eli eames x
ER LOLY GENS Spo teens ause sree I SiMe Redes sn CRSA dua Pee aA SH ee Te i
Arthropoda :
Phillipsia tuberculata M. & W.| x EX, ai | Prcaenea sy eeges conn teeny Su iodine eases Oca Heer Aa
Griffithides sedaliensis Vogdes..| x x Gl [Fa aeece Bh Feed eet a ete Ree
above have been described from higher formations only, but as
yet the exact range of the Mississippian fenestelloid species is
imperfectly known, and such species might well be represented in
the earlier faunas. The presence of such fossils as Centronelloidea
rowleyt, Rowleyella fabulites, Nucleospira obesa, and Ambocoelia
levicula, which are characteristic of the white chert division
of the Lower Burlington at Louisiana, Missouri, together
with the great assemblage of Lower Burlington forms, would
seem to indicate that this was the proper correlation for these
faunas.
570 JOSIAH BRIDGE
Bowlder No. 16, Locality 10, has yielded but two forms—
Syringothyris platypleurus Weller, and Conocardium sp. nov. The
former is a typical Lower Burlington form, and the latter is identical
with forms described from the bowlders referred to the Burlington
white chert. While it is not possible to place such a small fauna
definitely, such evidence as there is seems to indicate that it should
be correlated with the bowlders which have just been described.
These faunas, with the exception of the one from bowlder No. 27,
were obtained from bowlders of sandstone or quartzite. Bowlder
No. 27 was composed of white chert. As a general rule, however,
it may be stated that the Burlington faunas are to be looked for
in the quartzite bowlders, and the younger and older faunas in the
cherts.
A number of other bowlders have yielded faunas which are
apparently Lower Burlington in age, but which do not contain a
fauna which may be classed as being truly representative. These
faunas are listed below:
LocaLtity 10o—BOWLDER No. 1
Platycrinus sp.
Cliothyridina tenuilineata (Rowley)
LocALITY 10—BOWLDER NO. 6
Productus sp.
Rhipidomella diminutiva Rowley
Terebratuloid shell gen. ? sp. ?
Spirifer biplicoides Weller
Spirifer gregeri Weller?
Spirifer sp.
Locality 12—BOWLDER No. 19
Fenestella compressa var. elongata Cumings?
Brachythyrts sp.
Phillipsia tuberculata M. & W.
Locality 7—BOWLDER NO. 21
Spirifer platynotus Weller
Spirifer rowley Weller?
Spirifer sp. nov.
Unidentified crinoid
The largest fauna which has thus far been obtained has come
from a large chert bowlder which was found by Professor Dake.
MISSISSIPPIAN OF PHELPS COUNTY, MISSOURI 571
This bowlder is so completely weathered that the original texture
is entirely gone, and in its place is a porous siliceous mass which
powders and crumbles and breaks irregularly. The original bedding
of the formation to which this bowlder belonged is well preserved
in the specimen and is indicated by a slight lithologic change and a
more marked faunal change. Crinoid fragments and pieces of
bryozoans are to be found in all parts of the mass, but the former
are much more abundant on one side and the latter on the other,
and the dividing line is plainly marked. At first an attempt was
made to keep the faunas of the two sides separate, but it was soon
found that specimens of practically all the species represented were
found on both sides of the bowlder. The two faunas are therefore
treated as one.
FauNA OF BOWLDER 102—LOCALITY 1
Coelenterata
Amplexus fragilis White & St. John
Monolipora beechert Grabau
Gen. ? sp.?
Echinodermata
Codaster sp. nov.
Platycrinus hemisphericus M. & W.
Platycrinus pratteni Worthen ?
Platycrinus sp., at least five species, probably more
Dichocrinus scitulus Hall ?
Dichocrinus sp.
Fragments
Molluscoidea
Fenestella cestriensis Ulr. ?
Fenestella cingulata Ulr.
Fenestella compressa Ulr.
Fenestella compressa var. elongata Cumings
Fenestella compressa var. nododorsalis Ulr.
Fenestella exigua Ulr. ?
*Fenestella filistriata Ulr.
*Fenestella filistriata Ulr. ?
Fenestella funicula Ulr.
Fenestella limitaris Ulr.
Fenestella multispinosa Ulr.
Fenestella multispinosa Ulr. ?
Fenestella regalis Ulr.
Fenestella rudis Ulr.
572 JOSIAH BRIDGE
Fenestella tenuissima Cumings
Fenestella triserialis Ulr.
Fenestella triserialis Ulr. ?
Cystodictya lineata Ulr.
*Polypora burlingtonensis Ulr.
Polypora gracilis Prout
Polypora hallana Prout
Polypora maccoyana Ulr.
Polypora radialis Ulr.
Pinnatopora conferta Ulr.
Hemitrypa perstriata Ulr. ?
Rhombopora dichotoma Ulr. ?
Rhombopora incrassata. Ulr. ?
*Schellwienella burlingtonensis Weller
*Productus parvulus Winchell
{Productus ovatus Hall
Productus mesalis Hall?
Pustula biseriatus (Hall)
*Rhipidomella diminutiva Rowley
*Schizophoria subelliptica (W. & W.)
Tetracamera missouriensis Weller
Rhynchopora beechert Greger
*Dielasma burlingtonensis (White)
*Cyrtina burlingtonensis Rowley
Cyrtina neogenes Hall and Clarke
*Spirifernia solidirostris White
Spiriferina norwoodana (Hall)
Delthyris similis Weller
Spirifer rostellatus Hall
Spirifer logani Hall
Spirifer tenuimarginatus Hall ?
Spirifer tenuicostatus Hall
{Brachythyris suborbicularis Hall
Pseudosyrinx keokuk Weller
Reticularia pseudolineata (Hall)
Eumetria verneuiliana Hall
*Nucleospira obesa Rowley
*Cliothyridina incrassata (Hall)
Cliothridina parvirostris (M. & W.)
Mollusca
Cypricardinia sp.
Orthonychia pabulocrinus (Owen)
Platyceras obliquus (Keyes)
Platyceras equilateralis Hall
Undetermined forms
MISSISSIPPIAN OF PHELPS COUNTY, MISSOURI 573
Arthropoda
*Phillipsia tuberculata M. & W.
Griffithides sedaliensis Vogdes ;
A careful comparison of this fauna with the extensive collection
in the Walker Museum, from the Keokuk and Warsaw formations
from a number of localities in the Mississippi Valley, has indicated
the approximate age of the fauna, but does not allow it to be
placed exactly. Its affinities are with the Upper Keokuk and
Lower Warsaw. What little evidence there is seems to favor the
placing of this bowlder in the Lower Warsaw.
Spiriferina norwoodana, Pustula biseriatus, Delthyris similis,
Fenestella exigua, F. funicula, F. compressa var. nododorsalis, Poly-
pora gracilis, and Pinnatopora conferta occur in the Lower Warsaw
at a number of localities. Eumetria verneuiliana, Pseudosyrinx
keokuk, and Monolipora beechert, while found in both formations,
appear to be more abundant in the Lower Warsaw. On the other
hand, Fenestella rudis, F. regalis, Polypora hallana, and P. maccoyana
have heretofore been identified from the Keokuk only. Spirifer
rostellatus is well known in the Keokuk, but its presence in the War-
saw has been regarded as doubtful.
An interesting recurrent Burlington association appears in this
bowlder, the species of which are indicated in the faunal list by an
asterisk. A few other species which are common to many Mis-
sissippian formations are indicated by the dagger. This recurrent
element corresponds almost exactly to the faunas assigned to the
Burlington white chert. It is known at the present time that such
a recurrent group of species does occur either in the top of the
Keokuk formation or in the base of the Warsaw, in southeastern
Missouri, and this bowlder was undoubtedly from this horizon
originally.
Closely related to the fauna just described are the three faunas
listed below:
BOWLDER 13—LOCALITY 10
Archimedes owenanus Hall
MissourI BUREAU OF GEOLOGY AND MINES, COLLECTION No. 4, LOCALItTy 12
Archimedes grandis Ulrich
Orthotetes keokuk (Hall)
Productus ovatus Hall
574 JOSIAH BRIDGE
BOWLDER 34—LOCALITY 6
Fenestella rudis Ulrich?
Archimedes grandis Ulrich
Productus ovatus Hall
Productus setigerus Hall
Rhynchopora beecheri Greger
Pseudosyrinx keokuk Weller
Camarophoria? sp.
The exact correlation of these faunas is a difficult matter, since
nearly all of the species represented are common to both the Keokuk
and the Lower Warsaw. Archimedes owenanus is a common
Keokuk form, but it also appears in the Lower Warsaw. A.
grandis is less common, but its stratigraphic range is believed to be
about the same as that of A. owenanus. A. wortheni, which is so
characteristic of the Warsaw faunas, has not been found. These
faunas are probably best classed as Keokuk. No trace of the
recurrent fauna mentioned on page 573, has been found in any of
them.
Missouri BUREAU OF GEOLOGY AND MINES, COLLECTION NO. 5
Fenestella serratula Ulrich
Fenestella serratula var.
Fenestella serratula Ulrich ?
Fenestella exigua Ulrich
Fenestella cf. exigua Ulrich
Fenestella exigua Ulrich ?
Fenestella sp. (3)
Polypora varsoviensis Prout
Polypora striata Cumings
Polypora spininodata Ulrich
Cystodictya lineata Ulrich
Hemitrypa sp.
Schizophoria sp.
Cypricardinia sp.
This is apparently the youngest fauna which has been found
thus far. It is characterized by Polypora varsoviensis, the large
forms of which are found in great abundance in almost every
fragment. This species is common in the Warsaw and younger
beds. The specimens referred to Cypricardina sp. do not appear
to belong to the same species as the ones found in bowlder No. 102,
MISSISSIPPIAN OF PHELPS COUNTY, MISSOURI 575
but are closely related. This fauna is probably Warsaw, but, in
the absence of detailed faunal studies of the Warsaw, a more exact
correlation cannot be made.
Briefly summarized, the results of this study show that the faunas
obtained from these residual bowlders are much more diverse than
has previously been supposed. They indicate a partial sub-
mergence, in early Mississippian time, of a considerable portion of
the northern end of the Ozark uplift. From the study of so small
an area, not much is to be inferred as to the movements and dis-
tribution of the Mississippian seas during the time when the forma-
tions represented in these bowlders were deposited.
Typical Chouteau faunas and pre-Chouteau faunas are con-
spicuously absent, though farther north the Chouteau formation
is represented by small outliers and scattered patches of bowlders.
From this it would seem as if the sea did not cover this part of the
uplift during Kinderhook time, but that a gradual submergence
during the late Kinderhook allowed the Burlington seas to invade
this area. This supposition is further strengthened by the sandy
character of many of the bowlders, which suggests that the ancient
shore line was not far distant. An alternative view is that the
Kinderhook formations, or part of them, were present, and, not
‘being as resistant as the younger formations, have entirely dis-
appeared. However, if this is the case, why should the Chouteau
formations occur as bowlder deposits in the counties between this
area and the Missouri River? On the whole, the evidence seems
to favor the first hypothesis.
Not much evidence of the Upper Burlington with its typical
crinoid fauna has yet been found, but some specimens in the
original School of Mines collection suggest that it was represented.
In most places where it is exposed at present it contains much less
chert than does the Lower Burlington, and this fact may account
for its failure to be more commonly preserved. The Keokuk and
the Lower Warsaw are probably both represented, but so great is
the similarity between their faunas that the few collections which
have been obtained do not suffice to make the distinctions clear.
Up to the present time, no evidence has been obtained of any
faunas younger than the Lower Warsaw.
EVIDENCE BEARING ON A POSSIBLE NORTHEASTWARD
EXTENSION OF MISSISSIPPIAN SEAS IN ILLINOIS
W. W. DAVIS
University of Chicago
CONTENTS
INTRODUCTION
LOCATION OF THE BOWLDERS
LITHOLOGIC CHARACTER OF THE BOWLDERS
PHYSICAL CONDITION OF THE BOWLDERS
CRITERIA BY WHICH THE BOWLDERS MAY BE RECOGNIZED
THE FAUNA OF THE BOWLDERS
THE GEOGRAPHIC AND GEOLOGIC RELATIONSHIPS OF THE FAUNA
THE PLACE OF ORIGIN OF THE BOWLDERS
CONCLUSIONS
Introduction.—F or some time it has been known that fossils of
Mississippian age can be collected within the city limits of Chicago.
These Mississippian fossils, which occur in bowlders in the glacial
drift, were brought to notice by Mr. William Johnston, who
reported them to Professor Weller in 1915. The collections which
form the basis of the present report are-.in part those first secured
by Mr. Johnston; others have been collected by Professor Weller,
and still others by the writer.
Location of the bowlders.—The drift bowlders which have afforded
the fossils occur in the southeastern portion of the clay-pit of the
Carey Brick Company, located near the northeastern corner of
Grand Avenue and New England Avenue, in the northwestern
portion of the city, between Hanson Park and Montclare, being
closer to the latter place. The clay-pit is excavated in a terminal
moraine, which belongs either to the lake border or to the Valparaiso
morainic system, but probably to the former.‘ The morainic
t Areal Geology Sheet, Chicago Folio.
576
MISSISSIPPIAN SEAS IN ILLINOIS 577
material contains a considerable number of glacial bowlders of
various types and sizes. Some of them are of igneous origin, others
carry fossils of Niagaran age. Those containing the Mississippian
fauna were observed only in the southeastern part of the pit. The
bowlders have been piled into many small heaps by the workmen,
several of which are made up almost exclusively of those of Missis-
sippian age. The collections which have been made were largely
secured from these piles, but similar bowlders may be found
scattered over the southeastern part of the floor of the pit.
Lithologic character of the bowlders.—In their lithologic character
these Mississippian bowlders are dolomitic limestones, which
closely resemble the Niagaran dolomite of the Chicago region in
general appearance; when fresh they are bluish gray in color and
very hard, but on weathering they take on an earthy, yellowish-
brown color and become very soft.
Physical condition of the bowlders——The great majority of the
bowlders have well-weathered surfaces, and in many examples the
outer surface to the depth of an inch or more is decomposed to a
soft, yellowish, more or less porous rock as a result of weathering.
Some large pieces are completely weathered to the center. This
weathered condition of the Mississippian bowlders is, in general,
in marked contrast to that of the bowlders of other ages found in the
pit. Many of the Mississippian bowlders are rather angular, and
none have well-worn faces; they vary in size from small fragments
up to irregular blocks containing two or more cubic feet; in general,
they lack the characteristic appearance of typical glacial bowlders.
If their location were not known, they might easily be mistaken for
residual fragments of weathering.
So far as now known, these bowlders are confined to the south-
eastern section of the Carey clay-pit. A few small masses with the
same lithological character were found on the surface of the moraine
near the pit, but as none of these fragments have afforded determin-
able fossils they cannot be identified with certainty.
Criteria by which the bowlders may be recognized.—The Mis-
sissippian drift bowlders commonly may be recognized by their
dirty, yellowish, weathered surfaces and by their two most marked
paleontological characteristics, namely, (a) an abundance of
578 W. W. DAVIS
crinoidal remains, mostly stems, and (0) the presence of numerous
specimens of Spirifer with plicated fold and sinus.
The fauna of the bowlders.—These bowlders are abundantly
fossiliferous, but the fossils are largely fragmentary and are com-
monly poorly preserved. Unfortunately they are nearly all in the
form of molds, a condition of preservation which has added to the
difficulties of their identification. None of the fossils are silicified,
but a few of the cavities left by the solution of the shells are
sprinkled with crystals of pyrite and dolomite. The extremely
weathered portion of the rock is generally too soft to yield deter-
minable fossils, and the unweathered portions yield comparatively
few. The great majority of the better specimens have been
collected from the partially weathered portions.
The most abundant fossils are the crinoids, although most of
the specimens are mere fragments of stems, few of which can be
identified even generically, and they are of little scientific value
except to show that the rock has been a conspicuously crinoidal |
limestone. Next to the crinoids the brachiopods are the most
common fossils, and owing to their abundance and better state of
preservation they form the most satisfactory element of the whole
fauna. Besides the crinoids and the brachiopods the fauna con-
tains corals, blastoids, bryozoans, pelecypods, and gastropods,
none of which are represented by numerous species or specimens.
The composition of the fauna is shown by the accompanying list
of species that have been identified (Table I). The number of
examples of each species that has been observed is recorded after
each name for the purpose of showing the relative importance
of the several members of the fauna; the geologic range and
geographic distribution are shown in the several columns.
The geographic and geologic relationships of the fauna.—The
geographic and geologic relationships of this fauna are not difficult
to determine. Considering first the brachiopods, the list shows
that all the forms are found in the Mississippi Valley, although a
few species have a wider range. Of these brachiopods, two species,
Spirifer gregert and Spirifer mundulus, are Lower Burlington.
Three species, Dielasma burlingtonensis, Spirifer forbest, and
Spiriferella plena, are confined to the Burlington. Seven species,
579
MISSISSIPPIAN SEAS IN ILLINOIS
‘MOT}DE]JOD WnesnyL JoyeM “7 YFos3ouopy ‘Kaasng poI180j0ay a4njS sroumjy LEST uyayng “S'D'S A 2S UOIyNqI14sIp pue osuer 40 a
omo | = 3 : Pe col oon PPy
“VIM 2 ONY MM ¢ SHyoAvg spda9Kyo] g
5 “ i. Tee | eee es soho ¢sunbyqgo spsatqolg
ouo | x . es : (Oe est ea eee IleH $40) snyoyduong
epodosjsery
n = ; fia | Peo mo CC é ‘ds unip«évqIouny
_ : ; FeO o-oo OI OL é ‘ds unjragono10 Fy
epodAsajeg
¥ : 5, : 2 A ( Aacitaa| | ace Oa eee O POC e'ds s1akyqosurak §
x * * + is 2p eee eH 2Ma7d vpjosafinds
ee ¥ : a Tees | ees TeH ¢ Dy2ajsau Dpjada frag §
% Fe VO TS | eee ee “TOON tpevdgays dafrsig §
f s ite peace AQ[MOY SUNpUNU safrsig F
- z OTs | se ee IJIN 1498048 afraid ¢
- ie * I “Wd}VIG QW pOoomMION 2aqdof safisid §
* * ** * * OLS) | ee TPH 2207/pms v14044 021495
SC) Melee * - Zz s| + TREY Sesuauojsurjang vyjaumopidiyy
: . - _ go | eee ekcacrege ts eppeene aimee sce PUTA,
RQ IY MM YppI1Gxas varg soy
i = . = x . P | Fa pas (ON DOC OES, Ba d ‘ds XULAKSOpNAS J
és i m Oras IPIM, S2suauapsudaf snjonporg
B1}OIS PAON ‘SUTUIOA
‘ye ‘epeaoN ‘oorxeyy MoN ‘OIyO | x my - Ne e = Tee ees es sdipityd viojpun puanjidaT
BPePAIN | x A - z -OUYM ¢ Stsuauopsuriéng DUtsDjaIg
* = Te alse JIM ¢ Sunpradxaur ¢vyskyd
OOIXSTT MON | xy = a Tee eee MOTTEMS 4NOAg DULpLAKYION.D
aessouuay, | x ee a Bi z “TS TPAYOUIM, SHZSOIIZINUE S2JaU0YD
euvipuy |, am > n i 8 “7 TRE Szapjnaiqaogns siskyKyovag
adoiny
‘SuIUIOA MW ‘VPPAIN ‘OOTXTT
MON ‘OIyO ‘AxonyUay ‘euerpuy |, - ¥ z ; a 5 Goa | eee QT[eag’T Psopjaun, s1acyyy
epodorporig
cee eee eee quUsWIseIy ployjaysoue J
eozoAIg
. * < ; € |prewinys 2 UIMG O7aM snysv7qQ0]¢ 4)
Baplojse[_
OTS sae S]USWISRIJ o[qeyT}UepIUy
= = é : > ‘ a | cease OCC ECT ds SnUrsoDI
oorxey MON | : i Ge leat omens Sete ee ee preUur
NYS 32 UIMG S2dorIUN SNULAIKLOG
* ee a is I “**preumMnys ¢supunjos snurs203K21q
(suI9}$ JO JAISN]OXe) VapIoulID
Cie Sem soos ois xsl: obistie logis ‘(a]qegtuaprun) s[eI09d
SOreoo'y 19910 HNOS | stounyy | eaoy |ynyooy | TF | FID | gossorg
ee SSN |e me -Bulping| UwJ9,J : aaa
-WOAN uN:
NOILOGIALSIG: OIHd VAIO) SONVY OINO TOL)
STISSO,J NVIddISSISSI[. DAVIOLNOPT AHL £0 LSIT
580 W. W. DAVIS
Athyris lamellosa, Brachythyris suborbicularis, Leptaena analoga,
Ptychospira sexplicata, Rhipidomella burlingtonensis, Schizophoria
swallovt, and Chonetes multicostus, are not limited to the Burlington;
but the Lower Burlington seems to be the upper limit of Chonetes
multicostus, Lepvaena analoga, and Ptychospira sexplicata, while
Rhipidomella burlingtonensis and Schizophoria swallovi are not found
above the Burlington. Two species, Cliothyridina prouti and
Productus fernglenensis, are found typically in the Fern Glen, and
one species, S‘pirifer shephardi, is typical of the Pierson limestone
of southwestern Missouri, which may be the equivalent of the Fern
Glen. The Fern Glen commonly has been classed as uppermost
Kinderhook, but it is altogether probable that it should rather be
considered as lowermost Osage. Species of the two genera Pseudo-
syrinx and Syringothyris are found in the Burlington. One species,
Cyrtia inexpectans, which is very rare, has been described from
residual chert in Missouri, supposed to be of Keokuk age, and another
species, Spiriferella neglecta, is found in the Keokuk, but these two
species are the least certainly identified of any of those recorded.
The brachiopods of the foregoing list are clearly related to those
of the early Osage faunas of the Mississippi Valley, as these faunas
are developed in Iowa, Illinois, and Missouri, and the formation
from which they have originated may be certainly correlated as not
younger than the Burlington limestone, and in all probability as
Lower Burlington.
The other elements of the fauna confirm the correlation sug-
gested by the brachiopods. The abundance of crinoidal remains
immediately suggests the Burlington limestone. Both Diézygo-
crinus rotundus and Dorycrinus unicornis are typical Burlington
crinoids. Platycrinus is also found in the Burlington, though it is
not confined to that formation. The blastoid Cryptoblastus melo
is another member of the Burlington fauna and is quite limited
in its geologic range. The gastropod Euomphalus latus is another
characteristic Burlington species, and both species of Platyceras
recorded are reported from the Burlington limestone. The corals,
bryozoans, and pelecypods which have been recorded have not
been specifically identified, but all the genera recognized are known
to be present in the Burlington.
MISSISSIPPIAN SEAS IN ILLINOIS 581
The place of origin of the bowlders—As Montclare is 165 miles
in a northeasterly direction from the nearest known Mississippian
outcrop in the Mississippi Valley, the question at once arises:
From where were these Mississippian bowlders transported by the
glaciers’? The general direction of ice movement in the Chicago
region during the last glacial epoch was somewhat west of south,
the published records of the directions of glacial striae west of the
city showing directions varying from S.57° W. to due west.!
Considering the known direction of glacial ice movement, the
first inference is that the bowlders may have been transported from
the known Mississippian outcrops in Michigan, but a careful
consideration of the Mississippian formations of that state and of
their faunas makes such an origin highly improbable.
The Mississippian section of Michigan as given by Lane? has
been followed by all later writers.
In this section the following subdivisions are recognized:
Bayport or Maxville limestone: Light and bluish cherty
limestones and calcareous sandstones.
Michigan series: Dark or bluish limestones and dolomites, with
gypsum and blue or black shales; some reddish or greenish shales
and dark or red sandstones.
Marshall sandstone: White and red sandstone, often pyritic,
peanut conglomerates, sandy shales, whetstones, and red shales.
Coldwater series: Blue shales with nodules of iron carbonate,
sandstone, subordinate streaks of fine-grained limestone, black
shale at the base.
Berea sandstone: White sandstone, nowhere exposed in the
state.
The published list of fossils from these Michigan formations?
shows them to be faunally related to the Mississippian of Ohio
rather than to the Mississippi Valley formations. Writers are not
altogether in accordance regarding the correlation of these for-
mations, but, giving the widest possible latitude, the Montclare
bowlders must have been derived from a formation whose age was
* Chicago Geological Folio, pp. 506.
2 Annual Report, Geological Survey of Michigan, 1908, pp. 74-86. (1909.)
3 Geological Survey of Michigan, Vol. VII, Part II, pp. 253-70.
582 W. W. DAVIS
included within the time of deposition of the Coldwater, Marshall,
and Michigan formations of Michigan. The published faunal lists
from all these formations are made up largely of pelecypods, and
from all the available lists a single species, Spirifer forbesi (and this
identification is admitted to be very questionable), is recorded from
the Michigan series, which has been identified in the Montclare
collection. Even if the identification of this species from Michigan
is correct, it is still quite evident that the Montclare bowlder fauna
has no relationships with the Michigan Mississippian which are
worthy of consideration. Furthermore, the lithologic character
of the Montclare bowlders is totally different from that of any of the
Michigan formations.
From the foregoing consideration of the faunal and lithologic
characteristics of the Mississippian formations of Michigan it may
be assumed as a demonstrated fact that the Montclare bowlders
were not transported from across Lake Michigan to their present
resting-place.
The only alternative conclusion in regard to the source of these
bowlders is that they were originally in place at no great distance
from where they now lie. Such a conclusion is further confirmed
by the physical condition of the bowlders themselves. They are
angular in outline and exhibit much less wear than most of the
associated bowlders. The present weathered condition of the
bowlders is probably original weathering accomplished before they
were moved by the glacier. This is suggested by the different
degrees of weathering of the bowlders themselves and by the dif-
ferent surfaces of the same bowlders. Furthermore, the other
bowlders of similar composition which were buried with the
Mississippian bowlders in the glacial débris are still essentially
unweathered, even upon their surfaces.
Conclusions.—A study of these Montclare glacial bowlders of
Mississippian age seems to establish the following important
conclusions:
t. There existed, previous to the last glacial advance, an outlier
of Mississippian rocks in northeastern Illinois, probably resting on
limestone of Silurian age. The remnants of such an outlier may
well be in existence, completely buried at the present time by
MISSISSIPPIAN SEAS IN ILLINOIS 583
glacial drift. The actual position of this outlier cannot now be
determined, but it was at no great distance from the present position
of the bowlders and may well have been within the present limits
of Chicago.
2. The limestone comprising this Mississippian outlier con-
tained a prolific fauna of brachiopods, crinoids, and other forms of
life, every species of which, so far as they have been certainly
identified, is present also in the rocks of lower Osage age in the
Mississippi Valley. ‘The whole association of species suggests a
very definite correlation of these Mississippian rocks with the Lower
Burlington limestone of western Illinois, Iowa, and Missouri.
There is no suggestion whatever of any faunal connection with the
Mississippian formations of Michigan.
3. On the basis of the two foregoing conclusions, the extension
into northeastern Illinois of the Mississippian sea, which occupied
the Mississippi Valley region in early Osage time, may be assumed.
“SOME EFFECTS OF CAPILLARITY ON OIL ACCUMU-
LATION” BY A. W. McCOY'
DISCUSSION BY
C. W. WASHBURNE
New York
Students of petroleum are indebted to Mr. McCoy for the
three useful experiments in this paper. His first conclusion is that
oil may accumulate in the larger pores and other spaces of rock,
regardless of structure. This fact had been recorded previously in
the occurrence.of oil in lenticular sands, and I had shown its theoret-
ical necessity.” The additional experimental evidence is most wel-
come. ‘The main argument of the paper is an attempt to show that
capillary action possibly may lift the strata into anticlines. This
idea appears impossible for four reasons.
First, the pressures created by capillarity are exerted by fluids
in open spaces which communicate, more or less deviously, with
the ground surface. The perfection of this communication through
shale is of the same order as the perfection of transmission of the
capillary pressures (assuming that these exist) that are transmitted
through shale into sandstone. Moreover, any pressures in the
fluids in sandstones are exerted through all spaces, laterally as well
as vertically, and would be so equalized through the entire bed of
sand that only a local hydraulic gradient would be left to deform
the sand. That such slight difference of pressure is unable to tilt a
sand need not be argued. Even if the capillary pressures could
deform a sand, they would not disturb lower sands along the same
axis or lift any of the underlying strata.
Secondly, the amount of pressure available under Mr. McCoy’s
assumed conditions (p. 802) would not be ‘‘the difference in the
t Jour. Geol., XXIV (1916), 798-805.
2C. W. Washburne, Transactions A.I.M.E., L, 831.
584
EFFECTS OF CAPILLARITY ON OIL ACCUMULATION 585
capillary pressures of oil and water for that size of opening.” It
would be only (a) the pressure exerted by the water-oil surfaces
beneath the oil, less (6) the pressure of the oil-gas surfaces on
the hemiglobules of oil that are being forced out of the
small capillaries into larger spaces. Since (a) and (6) are of
similar magnitude, their difference, amounting only to a frac-
tion of an atmosphere, cannot be considered a cause of rock
deformation. |
In other words, Mr. McCoy takes a wrong basis for his calcula-
tion of capillary deforming force, and this nullifies the calculation
at the end of his paper.
Thirdly, the amount of capillary pressure exerted in a group
of tubes of variable size and having many lateral connections, as in
rock, cannot exceed the pressure in the largest of these tubes plus
the head required for an adjusting flow of gas or liquid through the
lateral connections from the finer to the larger pores. If shale is
cut by minute open joints or fissures, the low capillary pressures in
the latter must limit the effective capillary pressure per square
inch in all connecting pores.
Pores that are completely closed above the point of the highest
liquid-gas surface within them would feel the complete pressure
produced by that surface. Capillary action in completely inclosed
pores would not contribute to the general pressure or affect the
problem under discussion. In all connected pores, the general
capillary pressure could not rise appreciably above the mini-
mum capillary pressure, determined in the largest pores or
joints.
I believe that the second and third arguments reduce the prob-
able effective capillary pressure, under the conditions of Mr. Mc-
Coy’s problem, to a maximum of one or two atmospheres, if the
rock is cut by any minute open joints or large pores. Certainly
the pressure of 200 atmospheres, which he deduces, is out of the
question as a general capillary pressure in rock.
Fourthly, there is no definite orientation to capillary pressures
in rock-pores. ‘They push and pull every way and tend to balance
each other. The ideal distribution of water, gas, and oil assumed
by Mr. McCoy would cause a small capillary pressure of definite
586 C. W. WASHBURNE
orientation, but the assumed distribution is improbable. The
probable original distribution is a sand filled mostly with water
and imbedded in shale the pores of which are filled with water, oil,
and gas, without any regularity or order.
The three useful experiments so carefully described by Mr.
McCoy are worthy of careful study by students of oil, and I trust
that we may have more experiments from the same author.
PETROLOGICAL ABSTRACTS AND REVIEWS
ALBERT JOHANNSEN
Bowen, N. L. “The Crystallization of Haplobasaltic, Haplodi-
oritic and Related Magmas,” Amer. Jour. Sci., XL (1915),
161-85.
The terms haplobasaltic, haplodioritic, and so on, are applied by
the writer to simple (pure) artificial mixtures of feldspars of various
compositions, and diopside. He uses the rock names according to the
best and most recent practice. Various mixtures of these substances
were studied by the quenching method of thermal analysis, that is, by
sudden chilling at known temperatures, and the material was studied
microscopically. No attempts were made to study the optical proper-
ties of the minerals formed, it being necessary only to distinguish diopside
from plagioclase in these experiments, the results of which are plotted
in numerous diagrams. From his results the writer concluded that
crystallization controls differentiation in the subalkaline igneous rocks.
Bowen, N. L., and ANDERSEN, Otar. “The Binary System MgO-
SiO.,”’ Amer. Jour. Sci., XX XVII (1914), 487-500.
A study of equilibrium in the binary system MgO-SiO, by the method
of quenching. Forsterite (Mg.SiO,) and clinoenstatite (MgSiO,) were
found capable of existing in contact with liquid in the binary system.
Clinoenstatite was the only stable form of MgSiO, found. It has no
true melting-point, but at 1,557° breaks up into forsterite and liquid.
At 1,577° the forsterite dissolves. There is no eutectic between these .
two compounds.
Bowen, N. L. “The Ternary System Diopside-Forsterite-Silica,”’
Amer. Jour. Sci., XX XVIII (1914), 207-64.
An investigation of various mixtures of silica, calcium carbonate,
and magnesia. It was found that the systems diopside-silica and diop-
side-forsterite show the simple eutectic relations; forsterite-silica shows
one intermediate compound (clinoenstatite) unstable at its melting-
point; and clinoenstatite-diopside forms an unbroken series of solid
587
588 PETROLOGICAL ABSTRACTS AND REVIEWS
solutions, corresponding to the monoclinic pyroxenes. The triangular
diagram, therefore, shows only three boundary curves and one ternary
invariant point. The writer shows that crystallization may proceed
according to two different methods, and the importance of distinguishing
between them is discussed. The optical properties of the pyroxenes
are discussed at some length; extinction angles, refractive indices, and
optic axial angles are measured and the orientation is determined.
BowEN,N.L. “‘Crystallization-Differentiation in Silicate Liquids,”
Amer. Jour. Sci., XX XIX (1915), 175-91.
Laboratory experiments showed that olivine and pyroxene crystals
sink and tridymite floats in artificial melts of diopside, forsterite, and
silica. From the rate of sinking, the viscosities of the melts were found
to increase with increase in silica.
COLLINGBRIDGE, HARvEY. ‘‘The Determination of the Maximum
Extinction Angle, Optic Axial Angle, and Birefringence: in
Twinned Crystals of Monoclinic Pyroxenes in Thin Section
by the Becke Method,” Mineralog. Mag., XVII (109014),
147-49.
Gives a method for determining various optic properties by observa-
tions on twinned crystals which show the emergence of an optic axis
in one portion.
Cotimss, W. H. The Huronian Formations of Timiskaming
Region, Canada. Museum Bull. No. VIII, Geol. Surv.,
Dept. Mines, Canada. Ottawa, 1914. Pp. 27, figs. 3, pls. 1.
Cross, WHITMAN. Lavas of Hawaii and Their Relations. U.S.
Geol. Surv., Prof. Paper 88, Washington, 1915. Pp. 07,
map I, pls. 2, fig. 1, bibliography.
The writer describes, with considerable space devoted to the norms,
various olivine-bearing and olivine-free-, bronzite-, picrolitic-, nephelite-,
and melilite-nephelite-basalts, limburgites, soda-trachytes, trachyande-
sites, a kauaiite or oligoclase-augite-diorite, some basalt tuffs, and a
PETROLOGICAL ABSTRACTS AND REVIEWS 589
gabbro. Forty-three chemical analyses are given, and in most cases the
normative minerals are computed. In the general discussion the
characteristics, chemical compositions, normative compositions, the
relations of norms to modes, and the classification are considered.
Twenty-two pages are devoted to the distribution of the rocks in the
Hawaiian Islands and of analogous rocks elsewhere in the world. The
writer discusses the Atlantic and Pacific provinces, and the alkalic and
calcic series (alkali and alkali-lime series of Rosenbusch), and concludes
with a discussion of differentiation in the Hawaiian magmas.
Cross, Wuitman. “On Certain Points in Petrographic Classi-
fication,” Amer. Jour. Sci., XX XTX (1915), 657-61.
An answer to several criticisms of the C.I.P.W. system of rock
classification.
DatmER, K. Erlduterungen zur geologischen Spezialkarte des
Kénigreichs Sachsen. Sektion Treuen-Herlasgriin, Blatt 134.
2d ed. revised by E. Weise and A. Uhlemann. Leipzig, 1913.
Bp s58; pl:
Picrite, diabase, granite, quartz-porphyry, mica-porphyrite, various
contact metamorphosed schists, and sediments are described.
Daty, REGINALD A. Origin of the Iron Ores at Kiruna. Vetensk.
och Prakt. Underso6k. i Lappland. Stockholm, 1915. Pp.
35, figs. 4.
Expresses the view that the “inclusions”? of ore in the Kiruna
quartz-porphyry, are endogenous, and represent “frozen-in” units of
differentiation. The accumulation of such ore-masses by gravity is
thought to be the cause for the origin of the main ore bodies.
Daty, REGINALD A. Geology of the North American Cordillera at
the Forty-Ninth Parallel. Mem. 38, Dept. Mines. Ottawa,
1912. Pp: 840, maps 17, pls. 73, figs. 42.
Although the date on the title-page of this important memoir is
3.912, and the date of transmission 1910, it was not distributed until
‘IQA. In the meantime Daly’s Igneous Rocks and Their Origin, which
_ contains a much fuller statement of the theories expressed in chaps.
\
590 PETROLOGICAL ABSTRACTS AND REVIEWS
xxiv to xxviii, appeared. It is almost impossible in the space avail-
able here, to abstract a book of this character. The Table of Con-
tents alone covers 15 pages, and a synopsis given by the author 8
pages.
After describing the area covered, the author shows the various
subdivisions into which the Cordilleras have been divided, and suggests
various additions and changes. Then follow descriptions of the stratig-
raphy and structure of the Clarke, MacDonald, Galton, Purcell, and
Selkirk mountain systems, and the Rossland, Christina, Midway,
Okanagan, Hozomeen, Skagit, and other ranges. In chaps. ix to x the
Purcell lava and associated intrusives are described. The differentia-
tion in the Moyie sill is ascribed to the assimilation of quartzites, and
the writer offers proof of this as well as of gravitative differentiation.
A great number of chemical analyses are presented. The descriptions
of the rocks are given in a manner which might well be followed by other
petrologists, namely that of giving the mode of the rock as well as the
norm. Further, it is advisable, as is here done, to indicate whether
the mode was determined by the Rosiwal method, or by recalculation
of the analysis and comparison with the thin section. Pages 677 to 791
are mostly theoretical, and deal with the theory of igneous rocks, classi-
fication of igneous bodies, mechanics of batholithic intrusion, differen-
tiation, classification of magmas, etc.
The report is unusually interesting, not only in the theoretical part,
but also in the descriptive portions, which in most geologic reports have
a soporiferous effect.
Daty, Recinatp A. “Problems of the Pacific Islands,” Amer.
Jour. Sct., XLI (1916), 153-86, pl. 1, figs. 38.
A plea, given at the meeting of the American Association for the
Advancement of Science at San Francisco last August, for the establish-
ment of a central bureau for the comprehensive exploration, from a
scientific standpoint, of the Pacific Islands. It is estimated that the
cost of such a project will be from $800,000 to $3,000,000, depending
upon the thoroughness of the work, and that it will require about ten
years of time for the field work, and an additional five or ten years for
systematizing and publishing the results. The writer presents a number
of the problems which should be solved.
PETROLOGICAL ABSTRACTS AND REVIEWS 591
DRYSDALE, CHARLES W. Geology of Franklin Mining Camp,
British Columbia. Mem. 56, Geol. Surv., Dept. Mines.
Ottawa, 1915. Pp. 246, pls. 23, figs. 16, bibliography.
A report on a mining camp in the Yale District in south-central
British Columbia. The Franklin group contains the oldest rocks in the
district, consisting of metamorphic tuffs, quartzites, and argillites, the
latter carrying Paleozoic fossils. The rocks may represent early marine
coastal conditions of sedimentation and igneous activity prior to the
submergence and eastward transgression of a Carboniferous sea. At
the close of the Paleozoic the main folding and metamorphism of the
region took place, and the Franklin District thereafter remained above
the sea. During the Jurassic period there came the intrusion of a grano-
diorite batholith beneath a considerable cover of sediments. It did not
reach the surface. The Cretaceous period was one of long-continued
denudation, laying bare great thicknesses of Paleozoic rocks and even
exposing the underlying Jurassic batholith in places. At the close of the
Mesozoic the whole Cordillera was uplifted and the Valhalla granite
was probably intruded. The early Tertiary was a period of regional
sinking accompanied by some volcanic activity. It closed with the
tilting of the Kettle River formation, and a new cycle of erosion started.
At this time also there came the intrusion of monzonite. During the
Miocene there came intrusions of syenite, followed by pyroxenite and
augite-syenite, pulaskite-like dikes, and trachyte flows. Regional uplift
closed the Tertiary. During the Pleistocene all except a few of the
highest peaks of the Cariboo Range were covered by the Cordilleran ice
sheet.
A number of chemical analyses of the igneous rocks are given.
Eskoia, Pentti. On the Petrology of the Orydrvi Region in
Southwestern Finland. Bull. com. géol. Finlande, No. 4o.
Helsingtors,, 1914.) Ep. 277, pls’ 6, maps 2, figs..55, bib-
liography.
This interesting bulletin gives an account of the petrology of a series
of Archean metamorphic rocks in the vicinity of Orijaérvi. After a short
geologic history of the region, the author gives careful and detailed
petrographic descriptions of various granites, magmatites, pegmatites,
diorites, gabbros, hornblendites, aplites, peridotites, amphibolites,
leptites, and limestones. He then describes the exogenic contact-zones
of the oligoclase-granite, and gives petrographic determinations of the
592 PETROLOGICAL ABSTRACTS AND REVIEWS
cordierite-anthophyllite, quartz-cordierite-, cordierite-, and andalusite-
quartz-mica-rocks, cordierite-gneiss, plagioclase-biotite-gneiss, cumming-
tonite-amphibolite, and the skarn rocks. A great many chemical
analyses are given, and they ave recomputed into the norm as well as
into Osann’s system. Further, all analyzed rocks whose mode could
be determined under the microscope have been recomputed into the
mode,.an example which might well be followed by petrographers in this
country.
FENNER, CLARENCE N. “The Stability Relations of the Silica
Minerals,’ Amer. Jour Sci., XXXVI (1913), 331-84.
The following inversion-points were determined at atmospheric
pressure.
870°+ 10° quartzss tridymite
1470 + 10° tridymitesscristobalite
Velocity of transformation very slow.
a-quartz—>-quartz 575°
B-quartz>a-quartz 570°
a-tridymite>;-tridymite 117°
B,-tridymite>8,-tridymite 163°
a-cristobalite > -cristobalite 274° to 220°, depending upon the previous
heat treatment.
G-cristobalite>a-cristobalite 240° to 198°, depending upon the previous
heat treatment.
The transformation in the last six cases takes place promptly.
The melting-point of cristobalite is ca. 1,625°, while quartz is at least
155. lower.
Fermor, L. Lercu. ‘Preliminary Note on Garnet as a Geological
Barometer and on an Infra-Plutonic Zone in the Earth’s
Crust,” Records Geol. Surv., India, XLIII (1913), 41-47.
A comparison of the specific gravities of certain garnet-bearing rocks
with the specific gravities of the same magmas crystallizing in normal
minerals showed that the garnet-bearing rocks occupied from 10 to 20
per cent less room. From this the author concludes that garnet-bearing
rocks, such as kodurite, eclogite, etc., are high-pressure forms of normal
rocks. He therefore postulates the existence, below normal plutonic
rocks, of a shell characterized by garnets wherever a sesqui-oxide radicle
exists. For this shell he proposes the term ‘‘infra-plutonic.” Another
PETROLOGICAL ABSTRACTS AND REVIEWS 593
mineral of this zone is diamond. Under normal conditions the author
thinks a relief of pressure would liquify a certain portion of the infra-
plutonic rocks which, on being intruded into the higher zones of the
earth’s crust, would there solidify under less pressure as a normal plu-
tonic rock. Only under exceptional circumstances, for example when
the isogeotherms are lowered more rapidly than the pressure, will the
garnet-rock cool in its infra-plutonic form, to appear later by erosion.
The author considers this garnet-shell to be continuous around the
earth and potentially liquid, subject to local fusion and the formation of
reservoirs wherever there is a reduction of superincumbent pressure.
Applying this theory to meteorites, he thinks the chondrules, which
occur in so many stony varieties, were formerly garnets, and that clif-
tonite in the iron meteorites was formerly diamond.
FETTKE, CHARLES REINHARD. ‘“‘The Manhattan Schist of South-
eastern New York State and Its Associated Igneous Rocks,”
Ann. N.Y. Acad. Sci., XXIII (1914), 193-260, pls. 8, bibli-
ography.
The Manhattan schist, the youngest of the three crystalline meta-
morphic formations which form bed-rock in southeastern New York,
occurs in a series of closely folded anticlines and synclines, usually
unsymmetrical and in many cases overturned toward the west. The
axes of the folds run northeast and southwest and gently dip to the south.
The chemical composition and field-relations of the schist show that it
is of sedimentary origin, derived from shales, sandstones, and arkoses. .
These were laid down conformably upon the underlying limestone to a
depth of several thousand feet. Later a series of basic rocks—horn-
blende- and actinolite-schists of dioritic and gabbroic characteristics,
and granodiorite-gneiss (better gneissoid-granodiorite, since it was
determined to be of igneous origin)—was intruded in the form of sheets
and sills. Now came a period of intense folding accompanied by intru-
sions of granite, aplite, and pegmatite. Later there were intruded
various basic rocks—norites and pyroxenites of the Cortlandt series,
hornblendite near Croton Falls, and other rocks now altered to serpen-
tine. The pegmatitic intrusions still continued, for these later basic
rocks are cut by them in several places.
REVIEWS
The Origin of the Magmatic Sulfid Ores. By C. F. ToiLman, Jr.
and Austin F. Rocers. Leland Stanford Junior University
Publications, 1916. Pp. 76, figs. 7, pls. 20.
After reviewing the literature bearing on the modes of origin of the
various magmatic ore deposits, the authors have proposed as their
thesis that “the magmatic ores have in general been introduced at a
late magmatic stage as a result of mineralizers and that the ore minerals
replace the silicates. This replacement, however, differs from that
caused by destructive pneumatolytic or hydrothermal processes in that
quartz and secondary silicates are not formed at the time the ores are
deposited.”
The authors follow the position taken by Bowen in his recent work
establishing the process of fractional crystallization as the dominant one
during magmatic differentiation. After studying suites of specimens
from Sudbury, Ontario, Elkhorn, Montana, Ookiep, South Africa, and
Plumas County, California, the conclusion is reached that the ores have
been introduced by pneumatolytic means after the formation of the
rock-bearing silicates. The authors show clearly that the sulphides are a
late magmatic product, that they surround the silicates, cut them with
well-defined veinlets, embay them, and penetrate cleavage cracks and con-
tacts with other minerals. The absence of metallic silicates makes it
clear that the ores were not introduced as molten material, while the
replacement of early formed minerals indicates the presence of mineral-
izing solutions. The complete absence of reaction rims shows that the
replaced material was removed by the same agents which introduced
the ores. Selective replacement is shown by the preservation of the
original graphic texture of the rocks in the ores. There is also evidence
of the alteration of pyroxene to hornblende prior to the introduction of
the ore minerals, suggesting the presence of aqueous vapor. The small
amounts of hydrothermal alteration present appear to be related to a
post-magmatic stage.
The authors conclude that the temperatures involved in the deposi-
tion of the ores did not exceed 300 €. to 4oo C., but it is unsatisfactory
so to limit the temperature without further data than are here presented.
594
REVIEWS 595
vy
That the temperatures were higher than those of pegmatites is admitted
by the writers, but the only thing certain about the temperatures of the
pegmatites is that some minerals in some pegmatites formed at tempera-
tures lower than 575° C. The dominance of pyrrhotite as compared
with pyrite is recognized, and this indicates a high temperature, since
pyrite is less stable than pyrrhotite at such temperatures.
The paper represents an excellent piece of work and is a distinct
contribution to the knowledge of magmatic processes. It also serves
to emphasize the fact that a great deal must be known concerning the
minute textures of rock masses, other than that they are merely in juxta-
position, before positive conclusions may be reached as to the sequence
of crystallization.
E. A. STEPHENSON
CHICAGO
Relations of Cretaceous Formations to the Rocky Mountains in
Colorado and New Mexico. By Wiis T. LEE. Prof. Paper,
U.S. Geol. Surv. No. 95-C, 1915. Pp. 27-58, pl. 1, figs. 11.
In this paper physiographic principles are applied to certain phases
of the stratigraphy of the southern Rocky Mountains. The geographic
conditions during the Mesozoic are discussed and a large number of
Cretaceous sections are considered. This study indicates that this
basin of Cretaceous deposition was deepest in northern Colorado and
southern Wyoming, and that the main mass of sediment came from
an ancient land farther west. The sections show, moreover, that the
sandstone formations near this ancient continent become thinner east-
ward, toward the present Rocky Mountains, and are replaced by shales.
The author concludes that the conformable Cretaceous formations up
to and including the Laramie once extended across the present site of
the mountains. Downward warping and deposition in this basin was
followed by uplift and erosion. This change is believed to mark the
close of the Cretaceous. The formations deposited after the uplift
(the post-Laramie formations) belong in the Tertiary.
Laks 1885 15%
Review of the Pleistocene of Europe, Asia, and Northern Africa.
By HENRY FAIRFIELD OsBorN. Annals N.Y. Acad. Sci.,
XXVI, 1915, pp. 215-315; figs. 20, tables 4.
This paper is a revision for the German edition of chap. vi of the
author’s The Age of Mammals.
HRB.
596 REVIEWS
Geological Relations and Some Fossils of South Georgia. By J. W.
Grecory. Trans. Roy. Soc. Edin., L, 1915, pp. 817-22, pls. 2.
The outcropping rocks are described as consisting of a metamorphic
series of probable Ordovician or Silurian age and a series of marine
Mesozoic rocks associated with volcanic tuffs. There is no evidence
of Cenozoic volcanic activity, and the igneous rocks are not of distinctly
Andean types. Suess believed that the Andes extended in a great
horseshoe curve through South Georgia to the South Orkneys and
Graham Land.
H.R. B.
The Jaw of the Piltdown Man. By GERRETS. MILLER, JR. Smith.
Misc’ Coll, IGXV, No. 12, to15. Pp. 3a) pls:'5.
In 1912 the right half of an apelike jaw, a portion of a human brain
case, and other human bone fragments were found in a gravel pit at
Piltdown, Sussex, England, associated with an interglacial (early
Third?) fauna. Assuming that these remains represented parts of one
individual, Woodward established the genus Eoanthropus, characterized
by the combination in one skull of a human brain case and an ape-
like jaw.
Miller now compares casts of these fragments with specimens of
Pongidae and Hominidae in the National Museum and finds that the
brain case shows fundamental characters not known except in the genus
Homo, while the other fragments show equally diagnostic features
hitherto unknown except among members of the genus Pan (chim-
panzees). For the Pleistocene species represented by Woodward’s
Eoanthropus, Miller proposes the name Pan vetus.
ERB.
The Shinumo Quadrangle, Grand Canyon District, Arizona. By
L. FE. Noster.” US. ‘Geol. Surv., Bull. No! 540; tor43) Ep
TOO, pls.48, Mg. 1.
This bulletin presents the results of a detailed study of the western
part of the Kaibab division of the Grand Canyon. The section includes
rocks of Archean, Algonkian, Cambrian, Mississippian, and Pennsyl-
vanian age. The lower or Unkar group of the Grand Canyon Series
(Algonkian), in particular, is treated in considerable detail. The map ‘
which accompanies the bulletin represents the first detailed mapping
done in the Grand Canyon region.
ER ybe
THE PRINCIPLES OF STRATIGRAPHY
By AMADEUS W. GRABAU, S.M., S.D.
Professor of Palaeontology in Columbia University
“This monumental work will be of great value especially to the
younger generation of American Geologists, in broadening their
mental horizon and outlining the problems which rise for solution
in the twentieth century study of the rocks. It differs from other
manuals in the English language to such a dégree that it
supplements but does not supplant them.”—Science, July 14, 1914.
8vo, 1150 pages, 264 illustrations. Price $7.50
DESCRIPT.VE CIRCULAR SENT UPON hte
A. G. SEILER & =>, NEW YORK CITY
CONTRIBUTIONS FROM WALKER MUSEUM
A new series—volume II, No.1. The Structure and Relationship of Diplocaulus.
With plates. By Herman Douthitt. 28 cents postpaid.
TITLES IN VOLUME I
No. xz. The Vertebrates from the Permian Bone Structure of the-Fore Foot of the Dimetrodon.
Bed of Vermillion County, Illinois. With plates. By E. C. Case. 27 cents postpaid.
By E. C. Case. 53 cents postpaid. No. 7. Notes on Some Carboniferous Cochlio-
No. 2. Prodromites, A New Ammonite Genus donts. With descriptions of seven new species.
from the Lower Carboniferous. Withplates. By With plates. By E. B. Branson. 27 cents post-
James Perrin Smith and Stuart Weller. 27 cents paid.
postpaid. No. 8. The Osteology of Some American Per-
No. 3. Palaeontological Notes. With plates. — mian Vertebrates. By Samuel Wendell Williston.
By E. C. Case. 27 cents postpaid. ce cents ee 5 ;
ea at pap fae TESA toe tenon 0. 9. (1) The Osteology of Some American
Bee pen Net ae wattle Anown Be By EC. Permian Vertebrates. II. (2) Synopsis of the
Case. 27 cents postpaid. American Permo-Carboniferous Tetrapoda. By
No. 5. The Stokes Collection of Antarctic Fos- _ Samuel Wendell Williston. 29 cents postpaid. ,
NO-S ne Speen NOM TONY (CT) Atactocrinus, a New Crinoid Gents
sils.. With plates. By Stuart Weller. 27 cents from the Richmond of Illinois. (2) Description
postpaid. .. of a Ste. Genevieve Limestone Fauna from Monroe
No. 6. (1) The Osteology of the Skull of the County, Illinois. By Stuart Weller. 28 cents
Pelycosaurian Genus, Dimetrodon. (2) On the postpaid.
THE UNIVERSITY OF CHICAGO PRESS
CHICAGO 5825 Ellis Avenue - ILLINOIS
omeeoe
Specially Adapted to the
Teacher’s Requirements
The illustrated Model F-2 embodies
the quality characteristics of the
entire line of
Rausch [omb
Microscopes
Only our experience of more than 60 years as
makers of optical instruments enables us to offer
the teacher a Microscope which, he can feel con-
fident, will completely meet all of his require-
ments, optically and mechanically.
It has our Lever Adjustment with the necessary
slow movement for higher powers. It has ample
space for object manipulation and is thoroughly
practical and unusually durable. Price $31.50
Special Terms to Educational Institutions.
Many other Models illustrated and de-
scribed in our Catalog, sent free on request.
Bausch £9 Jomb Optical ©.
412 ST. PAUL STREET ROCHESTER, N.Y
New York Chicago Washington San Francisco
Leading American Makers of Photographic Lenses, Binoculars,
Reon eion Lanterns (Balopticons), and other High-Grade Optical
roducts.
Remington Junior
Typewriter
ON TEN DAYS’ FREE EXAMINATION
RETURN IT IF YOU WISH
The Junior is just a boiled down
Remington, smaller, lighter, and so simple
that anyone can operate it. No lessons
needed. 5
And it sells for $50. You can buy it on
easy terms—$10 down and $10 a month.
REMINGTON TYPEWRITER COMPAN
(Incorporated)
374 Broadway, New York
Send me a Remington Junior Typewriter, price $50, on free ex-
amination. It is understood that I may return the machine, if
I choose, within ten days. If I decide to purchase it, I agree to
pay forit in 5 monthly payments of $10 eacli.
FINE INKS 4x2 ADHESIVES
For those who KNOW
Drawing Inks
Eternal Writing Ink
Engrossing Ink
Taurine Mucilage
Photo Mounter Paste
Drawing Board Paste
Liquid Paste
Office Paste -
Vegetable Glue, Etc.
Are the Finest and Best Inks and Adhesives
Emancipate yourself from the use of corrosive and
ill-smelling inks and adhesives and adopt the
Higgins Inks and Adhesives. They willbea
revelation to you, they are so sweet, clean, well
put up, and withal so efficient.
At Dealers Generally
Higgins’
CHAS. M. HIGGINS & CO., Mfrs.
Branches: Chicago, London
271 Ninth Street Brooklyn, N.Y.
2a es
Are
So DS Resear Set eee nee es age
Ser aR eee
mn
ie
ee tee ee eee cee
RT Sa ee ae ee, a COEDS Re Ae ee
VOLUME XXV NUMBER 7
THE
JOURNAL oF GEOLOGY
A SEMI-QUARTERLY
EDITED By
THOMAS C. CHAMBERLIN AND ROLLIN D. SALISBURY
With the Active Collaboration of
SAMUEL W. WILLISTON, Vertebrate Paleontology ALBERT JOHANNSEN, Perrotocy
STUART WELLER, Invertebrate Paleontology ROLLIN T. CHAMBERLIN, Dynamic Geology
ALBERT D. BROKAW, Economic Geology
ASSOCIATE EDITORS
SIR ARCHIBALD GEIKIE, Great Britain JOSEPH P. IDDINGS, Washington, D.C.
CHARLES BARROIS, France JOHN C. BRANNER, Leland Stanford Junior University
ALBRECHT PENCK, Germany RICHARD A. F. PENROSE, Jr., Philadelphia, Pa.
HANS REUSCH, Notway WILLIAM H. HOBBS, University of Michigan
GERARD DrEGEER, Sweden
T. W. EDGEWORTH DAVID, Australia
BAILEY WILLIS, Leland Stanford Junior University
FRANK .D. ADAMS, McGill University
CHARLES K. LEITH, University of Wisconsin
GROVE K. GILBERT, Washington, D.C. WALLACE W. ATWOOD, Harvard University
CHARLES D. WALCOTT, Smithsonian Institution WILLIAM H. EMMONS, University of Minnesota
HENRY S. WILLIAMS, Cornell University ARTHUR L. DAY, Carnegie Institution
OCTOBER-NOVEMBER 1917
ON THE AMOUNT OF INTERNAL FRICTION DEVELOPED IN ROCKS DURING
DEFORMATION AND ON THE RELATIVE PLASTICITY. OF DIFFERENT
TYPES OF ROCKS - - . - = Frank D. Apams anp J. AusTEN BANCROFT 5097
ON THE MATHEMATICAL THEORY OF THE INTERNAL FRICTION AND LIMITING
STRENGTH OF ROCKS UNDER CONDITIONS OF STRESS EXISTING IN THE
INTERIOR OF THE EARTH - - - - - - Louis Vessot KInc 638
NOTE ON THE DEPOSITS CONTAINING oe REMAINS AND ARTIFACTS AT
ERO BEORTINA Mn ee rcrs (miei ag miniaelnes is ee i GEruanns Ose
' THE FOSSIL PLANTS FROM VERO, FLORIDA - _— - Beh Wie Epwarp W. Berry 661
. FURTHER STUDIES AT VERO, FLORIDA - - - - - Roriin T. CHAMBERLIN 667
ANOTHER LOCALITY OF EOCENE GLACIATION IN SOUTHERN COLORADO
WALLACE W. ATWwoop 684
ee ee ee ee ee
[EL BYD IIR IO 1E OTEVETC AGWIEC ANS ca Acs tea aka Meso a MTR SUTRA REI AN MP asa Mito a)
THE UNIVERSITY OF CHICAGO PRESS
CHICAGO, ILLINOIS, U.S.A.
g
THE CAMBRIDGE UNIVERSITY PRESS, Lonpon anp EDINBURGH
THE MARUZEN-KABUSHIKI-KAISHA, Tokyo, Osaka, Kyoto, Fukuoka, SENDAI
THE MISSION BOOK COMPANY, SHancuHatr
THE JOURNSE OF GEOloG.
EDITED BY
THOMAS C. CHAMBERLIN AND ROLLIN D. SALISBURY
With the Active Collaboration of
SAMUEL. W. WILLISTON ALBERT JOHANNSEN
Vertebrate Paleontology Petrology
STUART WELLER ROLLIN T. CHAMBERLIN
Invertebrate Paleontology Dynamic Geology
ALBERT D. BROKAW
Economic Geology
The Journal of Geology is published by the University of Chicago at the University Press semi-
quarterly, on or about the following dates: February 1, March 15, May 1, June 15, August 1, September 15,
November 1, December 15. {| The subscription price is $4.00 per year; the price of single copies is
6s cents. Orders for service of less than a half-year will be charged at the single-copy rate. {| Post’ te is
prepaid by the publishers on all orders from the United States, Mexico, Cuba, Porto Rico, Panama _ ..nal
Zone, Republic of Panama, Hawaiian Islands, Philippine Islands, Guam, Samoan Islands, Shanghai.
{ Postage is charged extra as follows: For Canada, 30 cents on annual subscriptions (total $4.30), on
single copies 4 cents (total 69 cents); for all other countries in the Postal Union, 53 cents on annual sub-
scriptions (total $4.53), on single copies 11 cents (total 76 cents.) {] Patrons are requested to make all
remittances payable to The University of Chicago Press in postal or express money orders or bank drafts.
The following publishing houses are authorized to quote the prices indicated:
For the British Empire: The Cambridge University Press, Fetter Lane, London, E.C. 4. Meany
subscriptions, including postage, 18s. 8d. each; single copies, including postage, 3s. 2d. each.
For Japan and Korea: The Maruzen-Kabushiki-Kaisha, ir to 16 Nihonbashi Tori Sanchome,
Tokyo, Japan. Yearly subscriptions, including postage, Yen 9.00 each; single copies, including
postage, Yen 1.50 each. ©
For China; The Mission Book Company, 18 Peking Road, Shanghai. Yearly subscriptions,
$4.00; single copies 65 cents, or their equivalents in Chinese money. Postage extra, if mailed direct
Steet of Shanghai, on yearly subscriptions 53 cents, on single copies rz cents.
Claims for missing numbers should be made within the month following the regular month “i publi-
cation. The publishers expect to supply missing numbers free only when losses have been sustained in
transit, and when the reserve stock will permit. Deliveries are not guaranteed within the war zones of
Europe, nor in other countries where transportation is disturbed by conditions incident to war.
Business correspondence should be addressed to The University of Chicago Press, Chicago, Il.
Communications for the editors and manuscripts should be addressed to the Editors of THE
JourNAL or Geoxocy, the University of Chicago, Chicago, Ill.
Contributors are requested to write scientific and proper names with particular care and in citations
to follow the form shown in the pages of the Journal.
Separates, if desired, must be ordered in advance of publication. Separates of leading articles will be
printed as ordered, of which 50 (without covers) will be furnished gratis, the actual cost of the remainder
(and covers, if desired) to be paid for by the author. The actual cost may vary, and will depend upon
the amount of work in remaking the pages into forms, presswork, paper, binding, etc. Separates con-
taining half-tones may be expected to cost somewhat more, the increase depending upon the number
of cuts and the amount of work required upon them.
Entered as second-class matter, March 20, 1893, at the Post-office at Chicago, Ill., under the Act of March 3, 1879
VOLUME XXV NUMBER 7
THE
POUNMINAL OF GEOLOGY
OCTOBER-NOVEMBER 1917
ON THE AMOUNT OF INTERNAL FRICTION DEVELOPED
IN ROCKS DURING DEFORMATION AND ON THE
RELATIVE PLASTICITY OF DIFFERENT TYPES OF
ROCKS
FRANK D. ADAMS, D.Sc., F.R.S., anp J. AUSTEN BANCROFT, M.A., Pu.D.
McGill University, Montreal
INTRODUCTION
At the meeting of the Geological Society of America held in
Albany in the year 1900, a brief résumé of the experimental work
on the flow of marble carried out by Adams and Nicolson was pre-
sented to the Society, and in the discussion which followed the
reading of this paper a number of interesting points were suggested
by various speakers as worthy of experimental investigation.
Among these was one put forward by Dr. G. K. Gilbert,
which, in a letter to the authors, he subsequently formulated as
follows:
It has been thought that great pressure breaks down the structure called
solidity and so reduces viscosity that very little differential stress is necessary
to produce flow. It is thought that the strength of rocks is practically un- -
affected by pressure, in which case flow should begin only when differential
stress equals the crushing strength of the material as conditioned by the
temperature. It is certainly conceivable also that the strength of rocks is
increased by pressure, so that the production of flow requires differential stress
greater than the crushing stress as conditioned by the temperature. I hope
your experimentation may be brought to throw light upon this point.
597
598 FRANK D. ADAMS AND J. AUSTEN BANCROFT
The sense in which certain terms are used in this quotation is
not quite clear, but we understand the question put forward by
Dr. Gilbert to be as follows:
A unit cube of any rock—granite for instance—is submitted to
pressure in a testing machine on the earth’s surface. It will give
away or break down under a certain load—this is termed its crush-
ing load.
If this cube of rock were imbedded deep within the earth’s
crust, great pressure would be exerted upon it from all sides. Such
being the case, and omitting from consideration the influence of
temperature, would the rock (1) be reduced to a condition which
approaches fluidity and move at once if the pressure in one direc-
tion became slightly greater than that in another? Or (2) would
the rock become deformed only when this additional pressure in
one direction was equal to its crushing load at the surface? Or
(3) would the rock show an increased resistance to deformation and
require a much greater additional pressure in one direction to
deform it than was required to crush it at the surface ?
A few preliminary trials which served to open up the experi-
mental investigation of this problem were undertaken some years
ago by Dr. Adams in association with Dr. Ernest G. Coker, then
Associate Professor of Civil Engineering at McGill University.
Dr. Coker subsequently resigned his position at McGill University
to accept the professorship of mechanical engineering and applied
mathematics at the Finsbury Technical College in London, and for
a time the work was discontinued. Dr. Bancroft, however, some
years later coming to McGill University, the investigation was
resumed. It has extended over a period of several years. The
writers desire to acknowledge their indebtedness to the Carnegie
Institute of Washington, the work having been carried out under
a grant received from that body.
ROCKS EXAMINED
The following rocks were examined:
White alabaster, Castelino, Italy.
White marble, Carrara, Italy.
Black Belgian marble (“Noir fin’’).
INTERNAL FRICTION IN ROCKS 599
White dolomite, Cockeysville, Maryland, U.S.A.
Steatite (“‘Albarine”’), Virginia, U.S.A.
Slate, New Rockland, Province of Quebec, Canada.
Sandstone, Cleveland, Ohio, U.S.A.
Granite, Baveno, Italy.
Olivine diabase, Sudbury, Province of Ontario, Canada.
For the purposes of comparison experiments were also con-
ducted with metallic copper and metallic lead.
Detailed petrographical descriptions of these rocks, with the
exception of the alabaster, dolomite, steatite, and slate, have been
given in a former paper.’ It is necessary here, therefore, to refer
briefly to the character of these four rocks only.
Alabaster, Castelino, Italy—Under the microscope the rock is
seen to be composed of an aggregate of small grains of gypsum
which are clear, colorless, and approximately equal in size. The
individual grains display a tendency to elongation in one direction,
thus giving the rock a very faint foliation. The columns of ala-
baster used in the experiments were cut from a single uniform block
of this rock in such a manner that their longer axes were parallel
to this indistinct foliation.
Dolomite, Cockeysville, Maryland, U.S.A .—This is a rather fine-
grained, white, granular dolomite, very pure in character and uni-
form in composition, containing CaCO, and MgCO, in almost
exactly their molecular proportions. It presents the appearance
of a white marble and is extensively quarried as such. Thin sec-
tions of the rock, when examined under the microscope, show that
it is composed of a mosaic of grains of the mineral dolomite, more
or less irregular in shape and varying somewhat in size. Between
crossed nicols, they present a uniform extinction or show only the
faintest strain shadows. They are very seldom twinned.
Steatite, Virginia, U.S.A.—This steatite is placed on the market
under the name of ‘‘albarine.” The columns employed in the
- experiments were cut from a perfectly uniform slab of this rock
«An Investigation into the Elastic Constants of Rocks More Especially with
Reference to Their Cubic Compressibility,”’ by F. D. Adams and E. G. Coker, The
Carnegie Institute of Washington, 1906; see also American Journal of Science, XXII
(August, 1906).
600 FRANK D. ADAMS AND J. AUSTEN BANCROFT
with dimensions of 10’’X11X14%"".. Under the microscope the
rock is seen to possess a distinct foliation parallel to the broad
surface of the slab. All of the columns were cut from this slab with
their longer axes parallel to the foliation. In thin sections under
the microscope the rock is seen to be composed chiefly of chlorite,
talc and dolomite, numerous small crystals and grains of magne-
tite, and a few grains of pyrite are also present. The two minerals,
chlorite and talc, make up by far the greater portion of the rock,
the chlorite being somewhat more abundant than the talc. Both
occur as plates and sheaflike aggregates, and both possess a very
distinct cleavage parallel to which extinction takes place. The
dolomite is present both in large rhombohedral individuals and as
small irregular granules which possess a linear arrangement parallel
to the foliation of the rock. None of the grains of dolomite show
either twinning or strain shadows. Having been cut parallel to the
foliation, it is not surprising that the columns of this rock employed
in the experiments bulged assymetrically when deformed, and
hence a larger number of experiments were made with the steatite
than with the other rocks, in order that accurate average results
might be secured.
Slate, New Rockland, Quebec, Canada.—This is a typical fine-
grained slate, black in color, uniform in character, and possessing
an excellent cleavage. By means of a diamond drill cores were
taken perpendicular to the cleavage of the slate, and from these
the columns of slate used in the experiments were prepared.
Under the microscope this slate is found to be composed essen-
tially of minute flakes of two minerals, one of which is apparently
kaolin and the other muscovite. In general, the kaolin is much
more abundant than the muscovite, from which it can be distin-
guished in that it possesses a lower double refraction and is not
quite so transparent. Within a few extremely narrow bands of the
slate the muscovite preponderates. A few minute grains of quartz
are interposed between the flakes-of muscovite and kaolin. A con- -
siderable number of very small flakes of black, opaque, carbonaceous
matter, abundant, minute, needle-like crystals of rutile, and a very
few widely scattered grains of pyrrhotite are also present. The
INTERNAL FRICTION IN ROCKS 601
rutile crystals are brownish in color and occasionally display the
geniculated twinning that is characteristic of this species.
The foliation of the slate explains the lack of symmetry in the
expansion of columns of this rock during deformation.
The Copper used in these experiments was taken from a rod
1 inch in diameter, representing a good commercial grade of this
metal. Prior to being turned into columns for the experiments, the -
pieces cut from the rod were annealed by being heated to bright
redness in the coal fire of a forge, being then allowed to cool down
gradually.
The Lead employed in the experiments was “‘assay lead”’ which,
in order to free it from all air bubbles, was melted down and cast
in a heated iron mold, which was then allowed to cool slowly.
METHODS EMPLOYED
Several long round bars of nickel steel 23 inches in diameter, all
of identical composition and from the same heat, and all having
been submitted to identical treatment in their manufacture, were
secured. For these the authors are indebted to the Bethlehem
Steel Company, which placed them at their disposal for the purpose
of the present investigation.
This steel, which is very uniform in character, possesses a high
tensile strength, as well as a high elastic limit, and has the following
chemical composition:
Carbone ere ute ele mre clrveenan wale .30 per cent
IMiameaniese Wier 2) Ai a, oP Nd .74 per cent
Silicomeeetirn nate rence trader. labia ters .162 per cent
Bhosphortuse mae cos a ere nie .035 per cent
SUMO MUI tee Alabseachs anesraiten enon: .038 per cent
INickelp py recs CMM aie calet adh ouiatk Soe hana 4.740 per cent
The bars were sawed into lengths of about 3% inches. These
were then bored and turned into tubes, the longitudinal sections
of which, with the final dimensions, are shown in the upper half
of Fig. 1. Two sets of these tubes were prepared, differing only in
the thickness of the wall of the central portion of the tube. In the
first set this has a thickness of 0.33 centimeter, while in the second
602 FRANK D. ADAMS AND J. AUSTEN BANCROFT
After Compression
Fic. 1.—Longitudinal section through steel cylinder with wall 0.33 cm. thick,
and inclosing one of the rock columns (natural scale).
INTERNAL FRICTION IN ROCKS 603
set the thickness is 0.25 centimeter. The interior diameter of the
tube in both sets is of such a size that it will just receive a column
of rock 2 centimeters in diameter. The inner surface of the tube
in every case was not only perfectly smooth, but highly polished.
The angle of the bevel, by which the thickness of the wall is reduced
at the middle of the tube, was adopted after a long series of pre-
liminary experiments, which proved it to be that which was
demanded by the conditions to be secured. Pistons fitting accu-
rately into either end of these tubes were then made of chromium
tungsten steel, suitably tempered by being heated, quenched in oil,
and then ground to the exact dimensions required.
Large blocks of each of the rocks having been secured, rough
columns of them were bored out by means of a hollow-bit diamond
drill, care being taken in the case of each rock to have all the
columns bored out of the rock in the same direction, that is, parallel
to one another, so that any possible variations due to rift, grain,
or incipient foliation were avoided. These rough columns were
then reduced to the exact size required, by being ground down in a
lathe by means of revolving carborundum wheels of different degrees
of fineness, and were finally highly polished. When completed the
columns were of such a size that they would just pass into the steel
tubes at the ordinary temperature, the tube inclosing the column
with an absolutely perfect mechanical fit. The column was in each
case 4 centimeters long and 2 centimeters in diameter. While the
column was thus fitted accurately into the tube, it could, by the
exertion of a certain amount of pressure, be moved up and down
within the tube. The column of rock, when inserted into the tube,
was so placed that its center was exactly in the center of the thinner
portion of the tube, as shown in the diagram, the extremities of the
column being in this way supported by the walls of the thicker
portion of the tube at either end.
The pressure to which the rock was submitted was obtained by
a Wicksteed testing machine set up in the Testing Laboratory of
the Macdonald Engineering Building of McGill University. This
machine has a capacity of 100 tons and, when loaded to its capacity,
is sensitive to a load of 4 pounds. Unfortunately, being graduated
to read only in tons and pounds, it was necessary to obtain the data
604 FRANK D. ADAMS AND J. AUSTEN BANCROFT
of the research in these units. In presenting the final results,
however, the data for the conversion of these into a unit more
generally employed in physical investigations are given.
The extensometer employed for the purpose of measuring the
expansion of the tube under pressure was a simplified form of the
type designed by Professor Coker and described in the Proceedings
of the Royal Society of Edinburgh, XXV (1904-5). It was affixed
to the opposite points of the steel tube on the plane of maximum
deformation and showed the expansion, multiplied by two, by
means of a fine line moving over a graduated scale, which was read
by a telescope placed at a distance of several feet.
In a number of experiments two extensometers were employed,
which were applied to the tube in the plane of maximum deforma-
tion, but in directions at right angles to one another. In this way
it was ascertained that the bulge which the steel tube displayed
under pressure was nearly symmetrical, but in order that any error
which might arise from a single measurement might be eliminated,
in almost all cases the two extensometers employed were affixed to
the tube at right angles to one another, and the mean of the two
readings was secured. By means of this form of extensometer and
by reading with a telescope, it was possible to measure-an increase
on the diameter of the tube amounting to only 0.0005 inch. The
steel tube inclosing the rock column, with the extensometers in
position, the whole set up in the press ready for the application of
pressure, is shown in Fig. 2.
The method adopted for measuring the internal friction devel-
oped in the rock by deformation was as follows:
A column of rock, Carrara marble, first was taken, having the
dimensions already referred to. This was inclosed in a tube of
nickel steel, as above described; the tube had a wall thickness of
o.25 centimeter at its thinner portion. As will be seen from Fig. 1,
the middle portion of the marble column is inclosed by the thinner
portion of the tube, while the ends of the column are held by the
thicker portion of the tube wall. In this way the rock is prevented
from flowing up between the tube and the pistons and thus from
escaping from the tube. With a tube of this shape and these
dimensions, the movement of the rock under pressure is confined
INTERNAL FRICTION IN ROCKS 605
to the middle portion of the column, which is surrounded by the
thinner portion of the tube. The pistons being inserted and the
whole properly set up in the testing machine, the pressure was
Fic. 2.—Steel cylinder, inclosing a rock column and with the two extensometers
in position, set up in the Wicksteed press. To the right a bulged cylinder is shown
as it appears at the close of an experiment.
applied in successive increments of 1,000 pounds. The exten-
someter showed no yielding of the inclosed rock until a load of about
12,000 pounds had been reached, when a very slight distension of
606 FRANK D. ADAMS AND J. AUSTEN BANCROFT
the tube was indicated. Up to this point, the marble, being
an elastic body, was undergoing cubic compression, the pressure
exerted by the machine and the resistance exerted by the steel
collar being equal. The slight distension of the steel tube at a load
of 12,000 pounds is due to the elastic deformation of the marble.
After each additional increase of 1,000 pounds to the load, exten-
someter readings were taken every 30 seconds until four successive
readings were identical, that is to say, until no movement that
could be registered on the scale took place during a period of
2 minutes. The pressure was then increased by another 1,000
pounds and a similar series of readings were taken. This was con-
tinued until the bulging steel tube showed signs of rupture or was
actually ruptured by the movement of the inclosed rock. The time
which elapsed between the first application of pressure and the
final rupture of the tube, that is to say, the duration of the experi-
ment, differed somewhat in the different experiments, but may be
said to be about four hours.
During the time which elapses from the point when the elastic
limit of the rock is exceeded to that at which the tube fails, the
inclosed rock is undergoing deformation with extreme slowness and
by internal movements of one kind or another, which give rise to
what may be termed a plastic flow.
At the commencement of the experiment the column of marble
had the form and dimensions represented in the upper half of Fig. 1.
When at the conclusion of the experiment the test piece was placed
in a lathe and the steel collar was turned off, the specimen of marble
was set free. It was still intact, unbroken, and, when tested in
compression, was found to be very nearly as strong as a piece of
the original marble of the same shape and size. It now had the
form represented in the lower half of Fig. 1.
A photograph of a column of rock, before and after deformation,
the rock, however, in this particular case being steatite, is shown
in Fig. 3.
The pressure which was applied to the marble column effected
two results. It overcame the pressure (or resistance) exerted upon
the sides of the column by the inclosing tube of steel, and it over-
came the internal friction developed within the rock during its
INTERNAL FRICTION IN ROCKS 607
change of shape. If it were possible, therefore, to ascertain the
amount of the pressure (or lateral resistance) exerted by the inclos-
ing tube, it would be possible by subtracting this from the total
load employed to determine the load which was required to over-
come the internal friction of the rock under the conditions of the
experiment.
In order to determine
the amount of pressure
required to effect the pro-
gressive deformation of
the tube, i.e., the amount
of pressure exerted by the
tube on the inclosed rock
during the successive
stages of deformation, a
series of steel tubes,
identical in every respect
with those employed in
the experiment just de-
scribed, were taken and : ee
° Fic. 3.—Photograph of columns of steatite
were deformed in a Pre- before and after deformation. The smaller
cisely similar manner, eX- _ divisions of the scale below are millimeters.
cept that these tubes were
filled with soft tallow, instead of being occupied by a column of
marble. This material was selected as being one which moves
with the development of an amount of internal friction which is
so small that it was negligible in the present case. In carrying
out the experiment with tallow, we found it necessary to slightly
alter the shape of the steel pistons, the ends inserted in the
steel tube being turned so as to present a somewhat concave face,
as shown in Fig. 4, the outer margins having a thin feather edge.
When pressure is brought to bear upon these pistons, this thin
edge expands slightly, thus pressing against the walls of the tube
and preventing the tallow from escaping between the piston and
the wall. It was found that in this way the deformation of the
tube could be readily effected.
608 FRANK D. ADAMS AND J. AUSTEN BANCROFT
The objection might be put forward that, while undoubtedly
the tallow possesses at ordinary atmospheric pressure an internal
friction which is quite negligible, this material under the pressure
to which it must be subjected in order to deform the steel tube
might develop an amount of internal friction and a rigidity which
would be by no means negligible.
In order to ascertain whether such was the case, companion
experiments were made, using the same pistons, but employing
ee
Fic. 4.—Longitudinal section through steel cylinder, showing the type of piston
used when deforming the steel with tallow.
water in one case and oil in another, instead of tallow. It was
found that the deformation of the tube could be effected by either
of these materials, although, when water was employed, it was
necessary to raise the pressure rapidly at first to cause the feather
edges of the pistons to expand and make the joint tight, thus pre-
venting the water from escaping. This series of comparative experi-
ments was carried out with loads up to 19,000 pounds, at which
pressure the tubes failed, and it was found that under these pres-
sures the three substances mentioned—water, oil and soft tallow—
showed no difference in viscosity which could be detected. The
tallow, of course, undoubtedly possesses a somewhat greater inter-
INTERNAL FRICTION IN ROCKS 609
nal friction than the water, but at the range of pressure to which
it was submitted in the present investigation this difference is not
noticeable and may therefore be neglected. The tallow, however,
being more convenient for purposes of experiment, was employed
in a further series of comparative experiments.
There was one other possible source of error, namely, the fric-
tion between the walls of the tube and the thin feather edge of the
hollow-faced piston used in the experiments with the tallow. In
the experiments with a column of rock a flat-faced piston was of
course employed, and this source of friction was thus eliminated.
In order to ascertain the amount of this friction in the case of the
tallow, another steel tube was constructed, identical in all respects
with those used in this investigation. One end of it, however, was
closed so that it would be necessary to employ only a single piston,
and through the closed end a small copper tube was inserted, which
led to a powerful pump provided with an accurate pressure gage.
The whole apparatus having been filled with water supplied by the
pump, the steel tube with its cup-shaped piston was placed in a
75-ton Emery testing machine, and the piston slowly forced into
the fluid, the pressure required to do this being noted at every
stage on the testing machine and also on the gage fitted to the
pump. In this way the pressure necessary to force the piston for-
ward was measured at each additional increment of load applied
to the piston by the Emery machine. As a result of a series of
trials, it was ascertained that the friction on the feather edges of
the piston amounted on an average to only 290 pounds, so that, in
view of the very heavy pressure employed in this investigation, the
error thus introduced is so small that it may be neglected.
It having been ascertained that soft tallow was a material which
for the purposes of this investigation might be considered to move
without the development of internal friction, a series of experiments
were made with steel tubes identical in character and dimensions
with those employed to inclose the marble, but soft tallow was
substituted for marble. The two series of experiments were
carried out in exactly the same manner in every detail, except that
in the tubes filled. with tallow the load was raised by increments
of 500 pounds, instead of 1000 pounds, and the readings were taken
610 FRANK D. ADAMS AND J. AUSTEN BANCROFT
every 15 seconds instead of every 30 seconds till they remained
constant for at least 5 consequent readings. This change was
necessitated in order to standardize the conditions in the two series
of experiments, since, when the tube was filled with tallow, the
whole load was applied to overcome the resistance of the tube,
while, when the place of the tallow was taken by marble, a portion
of the load was applied to overcome the internal friction of the
rock, and the movement was slower. By modifying the procedure,
as above mentioned, in the case of the tubes filled with tallow an
identical deformation was secured in both cases.
When columns of rock are inclosed in the steel tubes and
deformation is carried out in the manner described, the impending
rupture of the steel tube, which marks the conclusion of the experi-
ment, is indicated by the appearance of a series of sharply marked
vertical lines on the bulged wall of steel which inclosed the deformed
rock. If the experiment is continued, the tube splits along one of
these vertical lines, and the inclosed rock becomes visible, and, if
the pressure is still maintained, the resistance along the line of
rupture being removed, the rock along this line crumbles and is
forced out of the fissure in the form of a powder.
In the case of the experiments in which tallow was employed in
place of a column of rock, the completion of the test is marked by
the development of a vertical fissure in the thin portion of the steel
tube in the usual manner. So soon as this appears, however, and
usually before the load can be taken off the testing machine, a
fragment of. the thin steel wall, bounded on one side by the fissure
in question and at the top and bottom by the thicker portion of the
steel tube, opens out like a door on its hinges and is instantly torn
off and with a loud report is shot across the room with great violence.
It therefore was necessary in the case of these experiments that the
observer should always be protected from these projectiles, the
importance of this protection being emphasized in the case of one
of the experiments by the fact that the piece of steel struck and
split in two the piece of hard wood, a quarter of an inch thick,
which protected the observer’s head.
In order to make quite sure that the form and outline of the
bulge assumed by the tube in the case of the experiments with the
INTERNAL FRICTION IN ROCKS 611
different rocks was the same, a special series of experiments to
decide this question was made, employing copper, lead, marble,
Belgian black, and granite. In each instance the experiment was
carried to the point where the bulge or expansion of the diameter
amounted to 0.030. The cylinder was then removed, and by using
an electric arc light in a dark room a sharp shadow of the outline
of the bulged cylinder was cast upon sensitive paper, removed at
such a distance that the photograph enlarged the outline of the
cylinder approximately 18 times. The cylinder was then placed in
the Wicksteed machine, and the bulge increased to 0.110, and a
similar photograph taken. By a comparison of the photographs it
was found that the outline of the deformed wall was essentially
identical in all cases.
As has been mentioned, from two to five experiments were made
in the case of each rock when inclosed in a o.25-centimeter tube
and the same number with each rock inclosed in a tube having a
wall thickness of 0.33 centimeter. The mean of the closely con-
cordant results was then worked out in each case, and the figures
obtained are presented in Tables I and II. These represent the
data yielded by the experimental work.
The necessary data having been thus secured, a curve was
plotted presenting these graphically in the case of each experiment.
In these curves the exact amount of the load required to produce
any required bulge or distension of the tube is shown from the
point when the first movement can be detected until the final rup-
ture of the tube takes place. The curves for the several experiments
with Carrara marble inclosed in the steel tubes with a 0.25-
centimeter wall are shown in Fig. 5 (p. 620). A curve represent-
ing the mean of the results obtained in the several experiments is
also given. In Fig. 6 (p. 621) this curve of the mean of the results
obtained from the marble inclosed in a o.25-centimeter tube is
reproduced, and below it is the mean of the curves obtained from
tallow when inclosed in a 0. 25-centimeter steel tube.
Since the tallow, as has been shown, offers itself no measurable
resistance to deformation under the conditions of the experiment,
the curve in the tallow experiments shows merely the resistance
offered to deformation by the steel tube itself.
FRANK D. ADAMS AND J. AUSTEN BANCROFT
612
epeuey) | epeurd
Ayeqy ‘OUvIUO ‘Saqan(d)
‘ousaeg | ‘Ainqpns | ‘puepysoy
‘aytueiny | ‘aseqeiq: MON
SUIATI() 4R[S
eweasberaiiscety)|(sielseramac ate Sz00 Or00 g£0o SoIO 09z0 tees e ee ee ee le ee eee ee ee QQ olzz
erat ore ; Izoo 000 1f00 3900 eozO 0: 6)eF aM aiacientes|lesiel vile jeitel ewe 9: 0°0-0°0:0 60-40: OXDCyave pa tr4
PeCereCCet et th NiCr et tee Izoo 3000 9zo0o £00 £0z0 ein) (sete; (e.!m ce) |je vaio! (eujeue)eve\|isi\e) vie; es iaiin)-elelie **c¢00'0z
Cs oC Q10o S000 Iz0o gzoo SLIo Oh (CsDsOe cr On) Decettern cateo D0 -0=0 00 OOO Oeyeof(oNe
© Nele;e8,7e) 0,18) |,5 0 18|.040. ener es: g1oo S000 gi10o Izoo QviIo Os Os DOC Ox0a D4 boat COS csc at hcecmab aches ay CD “000‘QI
es ee ey g10o Sooo VIoo gi0o OzIO Oliéesi8s ine ttonel |(acoiellelsst.a serie OED ASDC ONO CHO Or alolo Fae
Bey aed a eR 6000 €000'0 | £100 £100 3600 CCsTeTaz Oe | jeegact eas ue | ee cocaeh acer OOOLO T
ease is erispialrel |Femnstemace ten ate 6000 steer ees! STO0 $100 ¢L00 E30 L6or‘0 Jt ttt ttt eQ00'Sr
Gfoeoeb=0 Gi cael icarecemroececed g000 se eeeee | OCF00 6000 oSoo oSSo: gblo ws) (elle isa vane. ence -*900'TI
w erecehstewede4| lence eteietsie 9000 abiremeris fe eajze 6000 6000 gfoo offo: Egto sieeijetisisemel lane enteiive “9000'S T
epkeea avrat iste mel laaWe jeitehercs, 'e goo0o evatiel she celle. © 6000 go0o Sz00 Ozzo Czfo: eifo) eelier ep enensite -*Q00'ZT
enioeke vetiesteite’|/a} sirens; oMeltetle Sooo eile stan sliswa ene 3000 go0o gI0o 610 gIzo° ster betes es SOOO TT
eit siieca\ wire Nedivi| fwiveMerreyenne avteo €000'0 |**''***:| Sooo gooo OI00 SLoo of to Freee eee ee 1 EQQOfOT
fe) oriole. e118, 6.04) |e ssixepielelfel ale fusiva|is\ eWewevene|teretarelarts joie £000 Sooo Sooo foo Z9Q00 ave! @ cease ee eheine. “0006
@,.6,.8 08; @.eite, Iie ele. a tee, 66/|1)6..0( 0)s0,.6:rsie ter] iel-«jq ke, ee ene £000 ooo £000 fe) OZ00 zFoo e220 810) dene a ejlenie “000g
04,8501... Orte. Oeil \c. 84,8 e) 6."e a. ethfe!ajle' 0. 16/678: :eltehe 6 eve, «)\6 6 [000 £000°0 @ faye) ene oxeixe: C100 Szoo DROID DONO U-Soveroyh
aeons jiabiiolet|fems.'s srejselre: sf {leis} -edeiesese nehlis 1s) a/célje)0 4 ele S000 sisigel is eas) ois iliane en eo sitetelis OIOCO QgIoo Tones sue OOOLQ
Seie}iensnwonene fo. |\eleyeqe:. asviece,||(e\'emehererese 16) | eee: es) stele C000 Sie erie ele. 0:0: isles (pirelce) ein Tooo'o CIO0O Deo OPT OS OUCOACD oye YayAy
# seirersefce).ey:eiiei||hectal je velverietei-ai|ijeine*loits, te e-:e' celt|%ev(s) eh tren ei ei= £000°0 acajten-anel ey, fe} |e of enelseL syaeitelieeWeie: ofp elise 000 OGposen BOG CA OO % yore} ar
eh ere sraie\lelreieui oer eins {onetneveiie!|| 8) rer.e <altetcns (whiel||{isi\a/fe/enlenvelas)ce||fetielvelraie strates |(anieitemerie setiecie)i|| vile! ai'ea slasiteriahlCsiysmaliatiamer cree: Sooo GeO O ne ORNONS AMOICAO Veoh AS:
Siero 97,8 limi :8) Lelie ahah fe e\4e 016s) (Asi e- ele (en) (el ales arte\jetiaine. siren] leet exe erie!teivet|lielieirelnitazisiteile,(Nal ecte,leljeltenief sil \eidei je: teteu si cene ZOO00'O D1 BeO 0B s0i 050.090 Oovoro pra
= te
Ne) poste n sentatt9 ‘Swo. eS ain
‘ayqie ‘purjaao Test (oe
GaAOTANY SOO
UALAWIINAD $z°O sO
SSANMOIBT, TIVA, V ONIAVH SYAGNITAD THALG AHL NI GaSOTONT NAHM SMOOY IVAGATS AHL LO ‘SHHONI NI ‘NOILVNAOIT f0 DNONONY
I ATaVeL
ise)
]
\o
INTERNAL FRICTION IN ROCKS
g100°0 | £¥00'°0
S100° £700"
V100° ooo’
£100° S£00°
I100° Itoo’
O100" Ifoo°
OI00° Qz00"
OI00" Szo0o°
O00" zoo"
O100" 0z0O0"
OI00° OzZ00°
Q000° QIOO°
3000 ° QIoo"
ooo" Q100°
ooo’ V1IOO°
ooo’ V10O'
ooo" V100°
ooo: V1o0o°
ooo’ VIOO°
ooo: 6000°
ooo 6000"
tooo 6000"
ooo 6000°
Vooo'o | gooo"
ic eaeen 9000"
Geer eee ee ooo .
eee eee we ooo .
© j0/6 ee ene. 6 ooo
Dine (e ie tallezane S000
0) ese le! e):p- 30) S000 fo)
O1z0'O
Sgio°
SfI0"
QorIo"
Sgoo°
£900°
£So00°
ooo:
gfloo°
££00°
ofo00°
zoo"
£z00°
£z00°
0z0o°
0z00"
Ozo00°
1lvo'o
govo-
1Sfo°
Iolo"
SSzo-
Vzzo°
IQIO°
€S1o0°
6z10°
Bolo’
1600°
SLoo:
0g00°
£So0o0°
gvoo°
Svo0o"
ooo’
ooo’
gtoo°
If0o0°
gI0o°
£100"
II0O"
I100°
‘ZI ‘SIqi JO UOISSNdSIp aq} UI pouTe[dxe se ‘uoUT
-19ads 94} 0} possAlpap JOVdUII UF SEM JIOY} SJUSWITIEdxa OY} JO 9UO UI ssNeVdaq IL[NSaII oe sjNsei ay} JoseqeiYy 0} Sururejziad uUNjoo 94} UT sIeZZep oy} Usemjog |
‘seraydsowje tg 6£1= your arenbs yod spunod 4°z$o‘z =] uuinjord ur usard se prot jo spunod ooo'r yor” »
06z0°
IZIO°
6g00°
goo:
6900°
0g00°
SSoo-
oSoo°
600°
SY00°
Ivoo°
(e}Xelons
goo"
£L00°
Ifo0o-
6z00°
gzoo°
gzoo°
vzoo°
£z00°
£z00°
IZOO0°
QIOO°
gIo0'o
Aye3y
‘oudAeg
‘ayueILDy
FRANK D. ADAMS AND J. AUSTEN BANCROFT
6£60 oy SU exenelieneserialze
6£80 ‘i ae |Uaiel eo ehe .
trlo On Ji llstevesemene .
oSgo Oo ko eceo eos O
SSo ei. on |teselieiretelcenieKe
SLto° 1zS1°o
g6£o° ceri
offo° QI
6gzo° €zz1
Lvzo° QIII
gozo° 1g60°
glio: So60°
6S10° TOgO
Qe10° foyefere)
Szio° obSo
€1I0° Lvvo
3600" Slo
6g00° oSto
goo" 6110
goo" QoIo
ooo: £o1o
gSoo° 1600
1S00° Sgo0o
gvoo'oO | gzoo'°o
epeue’) epeur)
‘ollejUCO ‘Jaqeng)
‘AINGpNG | ‘pueyyoy.
‘aseqriqd MIN
aura | 27%IS
roa 40) SPR fo) |]22.0°O Gop
ors FE ce eG
¢ 6 ¢ A pO) 221 pOeGao 0-0 7050
S1e1: ISQI‘o
fetace? 69S1°
€vi1 borr
flor oIti
og6o0 brer
3680 Sgz1
glo gAII
€zLo: O1II
$ggo° Ivor
£19g0° 3960
£vSo° S6g0
Sgto QIgo°
£10" Slo:
ogto° 1490"
€1fo0° ogo"
gSzo'o | gtSo'0
(,,Ug wsn
HION,,) |‘ purpAreyAy
‘aqqieN | ATTA
ueisjeg | -SsAayI09
PUL | ‘eywojoq
GaAOTINY SHOOY
‘oruO
‘purpAagD
‘QUOJSPUBS
Ajeqy
‘our[a}se_
‘Tayseqrlyy
‘vsn
“VIUISIT A
“9419894
ugddog
ava]
MOTIV I,
<SANNOg NI avo'T
614
panuyuo)—l ATAV.L
615
INTERNAL FRICTION IN ROCKS
“ULUI II
“Iy v
a
“urul OS
“Iy V
‘soraydsourze 9° 6E1 = your arenbs sod spunod 4° zSo'c =] UUMIOD UI UAAIS Se peo] JO spuNod ooo‘r WIV y
ulut VI | ‘uTW zo
“ry V ry v
z z
‘uIUL O | “UT O
my v “1y
v z
“UTU ZZ
“IY Z
is]
“UIUL TI | “Ul ze
“IY z@ “Ty I
v z
ulU OT | -uTUr zS
qf ae ‘Ty 0
z £
*-quount1odxa yoRa
jO UOI}VINP IsPIBAY
‘pouojiod s}usut
-Ladxo Jo TaquinNy
FRANK D. ADAMS AND J. AUSTEN BANCROFT
616
A[eI
‘OusARg
‘aquein
epeued
‘olIvjUugo
‘Ainqpns
‘aseqriq:
aulATIO
epeuey
‘Jaqoand
“purlyooy
MON
“a7eIS
‘aqruojoq.
aIAOTANY SIOOY
‘ys
HN 0)
‘puepaagD
‘Quo JSpuevs
ATei]
‘ourjayseg
‘Iaqseqely
Afseal
*BIUISITA,
‘9317893
adadog
ava]
MOTIVE,
xSANQ0g NI avo
YALAWIINAD ££'0 FO
SSENMOIHL, TIVM V ONIAV]] SUFANITAD THALG AHL NI GASOTONT NAHM SHOOY IVATAAS AHL AO ‘SAHON] NI ‘NOILYWUOTAG 40 LNAOWY
II ATaAVL
O17
INTERNAL FRICTION IN ROCKS
‘sazaydsourjze v9‘ 61 = yout axenbs sad spunod £°z$o‘z =] UUINjOD Ul UaAIS se PLOT Jo spuNod ooo‘r YoVE »
Vozo0'o
6L10°
$S1o0°
IfIo°
VIIO°
g600°
9800"
8900"
£g00°
tSoo°
tYyoo°
ooo’
If00"
Qzoo°
Qzoo°
Vzoo°
IZ0O°
IZ0O"
6100°
QI0o°
S100"
V100°
V1I0O"
£100"
ITIOO°
IIOO°
OI0O°
6000°
3000 °
go00°
ooo:
£000"
£000"
£000"
ofg0'0
0090"
gSSo-°
ofSo°
Sgto-°
gSvo:
gore’
€Lfo°
gSfo°
ozo"
oofo"
oLlzo°
gSzo°
ofzo°
OIzoO’
SgIo°
oLl10°
oSIo°
ovio'
QziI0°
QorIo'
£600°
$g00°
gLoo°
£g00°
oSoo°
ooo’
oboo’
of00"
Qz00°
£z00°
£z00°
OZ0O0"
0zOO°
o£30°0
€glo°
oblo:
$690°
gSgo°
O190°
oLSo°
SeSo:
£60"
zSto-°
Lavo:
6g£0°
6£o°
Sz£o°
g6z0°
£Lzo°
€vzo°
OIzO"
Sgro”
VoIo°
Ivio°
golIo'
300°
gSoo°
£v0o0°
QIoo*
£100°
£100°
II00°
OI00"
O100'
O00"
O100°
3000"
FRANK D. ADAMS AND J. AUSTEN BANCROFT
618
zloo 6610° 6090° Lvz1: o6zI QeLr‘o eee ee eee Cs ee ee euene.10. 0), vii[ieneia ove) iio: (eine je, 656) leiemesred| ie .eice, elke, (o\evleise) (e) 0 lus 000'zQ
Sgoo gLio° SzSo° POII’ 6zzL Sgor° aedetel clene. eh|fejvelelene) eo (ofi||ellelivierinjic) es eh lie vee, emeveeie'|e) es se;-0),:16 Uel.ei|/iet.c/i0, emeice ap len |) elenekelieitelm)se, 56) . 000‘1g
gSoo 610° gSfo° (Oya ree PLII Sor ore sels ofve'. 6-0; | 081.0] .0110,,.61) ep(8!| [leis en use! ferce,-e) 0, |Lejteue1e)- ee) 6 0)i|,e](e 0 yeriernel iene: |ise),es.e/ceesceie,iel|e<8)-emelleniv 620) .e:(0 1) eine 000'0g
SSoo 6F10° oofo' 6Sor' E11 €QSt eiieteNevsthiel ove4| esteloweimeceleney|ia owe ewednicenn eh ee) 01s ee) [ie .0: 10 Jejelne, ejne||\ceh.0..0/,0).0+'0\ 0) .0;|\uie)\emeiie salle) 19,0, \e) 18, <0 0006
zSoo 6z10° SlLzoO° Lior’ OgoL ovsti Syastzeltelibsiote.e.[:enjeizerielia; atte, jor|Veue,re- ‘eh ¢\lenese)l|lec.eie) erie ne! elsy|hereliol¢neles ehei|j20s.0\ tse! pete, (9,0) ,0)00)4einenelveyelce 58118 ue 000'9/,
gvoo C1I0O° Qzzo° 0960" CzoL Qgri ears SeLecce.e- vitel| |e; 0: faejaur terre ve: |ico.cer\en.6 cene> 6. lo" [hemes ol fejjerayis).vy |/e)zeiies\acvejianie)ia,|uienerne) ole) elieced| Xeon sriniprconeme(evseigenge 000‘ /,
©TO0O Soro’ CfIO° 0060: t960 IvV.I fe) fe Nverie soto Le jen) ny.el) |p uous venieel.etje||he) ones ocexeie!o) |holleuexale xe s0)| 2) (By ecsviene) je -ge)|\ge 10.70) ine (erie -fes |) ean elie ere nage else mas ae o000‘gl
ttoo g600" 0g00 ° brgo° 6060 oortr Sel crietlo erie) tal[Ualtellw we: (oftehelne} {ielkes.cs-'elieg eis? | seas ke eeted's))| lee calie colie.jsiel| siete eneeeljeme! |hegininelicdele:jeMenslis ms ae o00'S/
6£00 0600 ' gSoo° gogo’ gSg0 Coot eo asde We, (ee. we: |e. ele, 1e) (alcalie,le) [le syelefieltos elie iike) iwiiel ier ie/4mjlelis) | eiielietiene).e, iefie! | > sesimaienep {erro 95} 2 ule ehieisiie/ sel je uen.enie, 10 ood ts
1Eo0o0 £300" oSoo' ogo" O1go olzt Mice [aihetie uetteina||hetisiietteh’s Ste: fe) ali[lieiie/ke /e-.e2te Hel ei||lelpetleueile) elie, ei [latie is: te howheceus)| Memiveiypede-Totis\ger|{i> iim teuis)iat\sicalenie: le) iazee 000 FL
gtoo SLloo: oSoo° Z690° I9lo° CScr we ze Nel einer .b) er\|\eiiesce: jo. 'e, of ef lep||\«\famegiatie ater 'ed[holjetiaije,-vite! em eh [iene /¢. le) elieiiiel foi] estate Gane ainecde, |ietiswerieijegemegies eaealerur, ooo'z
goo 9900" gvoo- 690° 61lo° Sozi oSosro Cae aefel aC ey enl Vet esienlel.wmeheie:||\acenpraine tevne veyliel lene 06 eeillhw-iaceliminacie! iis heey, h sis Oy sire .t10e)Ne, 10 ooo rl
Sfoo gS$oo° 1voo° SgSo° IQQo° CSi1 g6rr: alremnevc ee, ele he lle 0) eivel-e cedip teliieieliegs! sifenef.s) fie: ee hegiai-elvs ail Os coLolwme lot oss ey ee ses) see let lane oo0‘ol
loo gSoo° ovoo: oSSo° Lovgo: Coll Srr1 eyes toc erierser « fhetehe tie; -oue elton |pel eg opreeirentei'e)] ee ibajhoeivlver,e)||teiiouel tecelectetias [io Thue, selmersemelse(ss! sstkormie 000'69
100 zSoo° 6£00° goSo- Legso SSol gLe1 © 104014 Jelle Le 10il|'e\,0:18¢ © (eo) "wilelte) [ceina) a, e710) ,0) 070) | a 10) en enerie,e,e)| #0, el. erielfeme.e cai|;oue'.4)0)Aep @ epee Le lege) 00°39
100 oSoo: SfCoo: 99vo° vVvSo CIOL oft eo 8 ea setae, eblie oe ake’ al 0, ie;\| eke! fecere: en'e,1e9|onese'-0)elbe ene! lad ete he tesco re Met |e, eke ie) ee Lee ete meme) ooo‘ Lg
8z00 Qboo° SSoo° vzvo: g6vo gf60 QLzI an pe eG ate a ihe \ele ea py e.uentie Nels) ne) le)iece | 18:0 le reine testeine Deiterie ju", gi ere} [le ieise jeuevie.«@ «10 (el sole 000'99
Lzoo CTO0O° ofo0o0' ogto* oSto: felelete) CZzl ele, alte le ite) [se sete yeme’ losis: e) [is esse 0) e. © 0i|)one.e-10) /e, {; o10)| io, 10.100 a 7x0) 00,0201) (0: eile! ie! 4) s0).et) el jelje) (a) re 000‘S9
gzoo ovoo: ofo00' 6£2£E0° o1tvo S9go gol miteyte dapie cos iei taille he @:(e,edea erin aielle te: rele le) |lareelieijei(erotlel|(iepietsyeiozvaie/celtel| su agla/hexa tpalenloniodoate Ke ooo‘ 9
Vzoo 9£oo° 6z00° Z1ISO* oLfo° Szgo° S111 atlotielfereilem.orig)|teltewrieneite ue fof [lei tei fe iepieine le © i|tei,eHelie: el leiyelier| ese vulbeccompire ge.|| keris ist elias [eihelnelieiiey tol te. ooo‘ £9
Vzoo vloo: Sz00° vlzo:° gffo° SgLlo' ogol el tetie) onto 'e, a ¢\| 6 Vesteselsesfere's [pal e ai’ 'a\ 0 00)||"eiteliay 0! je cele) py |ieetelie, fe, Jerte rie || ie) ¢ne see deyeno, @ 10 me pe) 000‘z9
CzZ00 ofoo" £Z00° LEzo° oofo gzlo- ZIOL aljaliaiehaie ace [Le-calsaee tae) stoe [facie O16) e..6.0bay|iel iat! elke nga teal Cvnel isco fee. Serta ity ii 8 Pe ge ee) 8 lel eles 900‘19
1zoo of00° OzO0O° vozo: 9Szo 0690" zS60 sisal folcac 6 a) 9 /ah|lelselt ave. (ftenteftp}]{e es ef este) er isliey |ieqimcre\ayias a oilwillie lepils; el leu eViniel (seu jeM(ejatverieye) (6) :4/i8) ve 00009
0z00'O 8z00'O 6100°0 SLIO‘'O Cozo fe) 3Sg0'0 4990°0 Citalieval-s’ aciet e:]iiav.civeleetor'erieicel|peljeel ie: (omenel,e:'| (e/le alee (svelte {miegren ehiel aise, pal eens ede) eenieliel eye 000‘ 6S
PUL ppeur: U UEC
er Gey ee ee | ae | cee || eee
OUDACE ‘Amapns “pueppoy ‘O[GTCTAL ‘ “OUT UE GUE nO soutjayse) “CIULAT A
oD | UTA. ons “BULL conmoredl ONAN |‘ouoyspuvg | MFVAV | ONIIS | ygagog | avery | MoTIvE, | ¥SaNO0d NI avoT
GIAOTANY SMOOY
panuywod—Il ATAVL
Oo
HH
Ko)
INTERNAL FRICTION IN ROCKS
“urut 6
sla
“UI O
MES)
‘uru FI | “Uru Ez | “Uru OF | ‘ur gz
eiued “ry v Iy Y ‘ay S
z Zz £ ¢
6rE1 re) one wliemwin ol @)-e6 [Se omel bet elelteil tee memeriese lenin
LECT: epee i [lsivetiece.ia).eete:| ie) «ehera tase, o)| ie seeeieteriete«
Qorr OS OwOstisGetieCacdy POs cic a-x orp) tethered rs
COTL Bijel.ajie xe) px elieh]irsieqefle/tele,.e'tey| helletetece ts:.e. 6
Loot 0) ol eie ele eels eceie arene ef|ic se heleha sae
L360 CORO vor CEC).| POO seer Un arl (O a Dsecata
9z60° 8 eon yelvehiel uiettle se ceselevese7.s4|\s lelsiece isha: s
vQgo° a2folfe OW er atererie' sae \enstesjee.i\< evel 0: eerie
3ogo LoviI fe) esen et siiere-ehen| vel sls lere cs) sr
6¢Lo oze1 crenejiv: ejtere- 4 |e»: 0, es vege eine
3490 €gz1 uretiale;lerievele! (cezenelp:o.eKens
6090" POIL eueieresie heey el iene: erene cacens
SLSo° OzII OS il lenotieweneuetelier |= .0)e.ilete7ere
sen Ze) SSor of ecu) ie) e) ane rallse® es -e.)) pra sel,
ozto ol60° eha/ elie Paviekexs: |eeires eetettelte
1Sfo° 630° S6vr1‘o | £vS1‘o
88z0° VIg0° Qtrr: glb1
¢Szo gtlo: ELe1 geri
£zzo SL9o0° g6z1 Soe1
“UIUL Qf
“ay S
‘ulul 7f
ayes:
“ULIN]OD YOOI SUIUIOJap ay} 0} PaJAtjap sem yovduat YSIS & par0jsor
Jo}v] UsyM PUL PIAOWAI SVM POT 9Y} SJUIUTIIOdxa oY} JO 9UO UI asneIaq Iv[NFoIII Iv S}[NseI ayy 9yWOTO, 0} Surureziad uUINoOD ay} UI sIaSSep ay} uaamyg |
‘soroydsourjye 9° 6£1= yout srenbs Jod spunod 4°zSo'c =] uumjogd ur uaats se peo] yo spunod ooo‘1 oe x
‘UTUL I
‘ry ¢
“urut £
"IY z
“UIUL OZ
“IY I
‘ulul V
‘IYI |**yuautedxa yore
jo uOl}eInp s8vIDAYy
z ‘pouroyjiod szusur
-ladxe Jo Jaquinyy
efoto nine teenie} | lexeuetiepere fe) etstlets 000‘Q0T
4\ifel eh tess lie-ertecBieiqer eller erie: elie eripiis o00‘ Lor
@/-a. iaice7 e-tevetel|\elieriamelie tal ea eres! 000‘QOT
Sice}ie!cevjeloGeiiel||e:rerlei elie Noises et elce felele Sor
‘sikei elie: lelss, teh ex|ivijer's telseijatielisi-asie\-16 ooo‘ tor
els) vevin (aljetiel-sE | scie el. shisj,eiceletistt eine ooo Sor
s) le\gaivesisiceueiie,{fategentereneserlesnaalis o000‘zOL
usesgeatelieltoseiied|ve te eiyecielvelellolitektepts: fefeley IOI
ORGS Cen Dactcy Honda Ocrh Oech tec 000‘OOrI
620 FRANK D. ADAMS AND J. AUSTEN BANCROFT
030 060 080 120 “150 180 ‘210 inches
Fic. 5.—Curves showing graphically the results obtained in four experiments on
the deformation of Carrara marble when it is inclosed in a steel cylinder with wall
o.25 cm. thick—also the mean of these curves (in heavy line).
INTERNAL FRICTION IN ROCKS 621
70,660
60,000
50,000
40,000
30,000
20,000
10.000
030 060 090 120 \ 150 180 ‘210 inches
Fic. 6.—The mean of the curves obtained in the deformation of Carrara marble
(see Fig. 5) when it is inclosed in a steel cylinder with wall o.25 cm. thick—with the
curve obtained when a steel tube of identical dimensions is deformed when filled with
tallow.
622 FRANK D. ADAMS AND J. AUSTEN BANCROFT
Such being the case, with the information thus secured it is
possible to separate the two components of the load, namely, that
necessary to overcome the resistance offered by the tube and that
required to effect the deformation of the marble. If at a series of
points the load required to produce a certain distension or bulge
in the steel tube when filled with the tallow is subtracted from the
load required to produce the same bulge in the case of the tube con-
taining the marble, values are obtained which represent that por-
tion of the load which is expended in affecting the deformation of
the marble. This may be termed the érue curve, and that obtained
for a standard column of Carrara marble deformed in a standard
steel tube having a wall thickness of 0.25 centimeter is shown in
Fig. 7. In the same manner the érue curve for each of the other
rocks may be plotted from the data presented in Tables I and II.
It will be seen that, in the case of Carrara marble, this curve start-
ing from a distension of 0.001, which may be considered to be due
to elastic deformation, and which is produced by a load of 12,000
pounds, shows a rapid deflection to a point representing a distension
of 0.052 which is produced by a load of 33,000 pounds, after which
it develops into what is practically a straight line until the tube
ruptures.
This shows that after the elastic limit of the marble has been
passed, at about 12,000 pounds, and the marble commences to
deform, the load which is required to start this movement and
produce a unit of diametral expansion is relatively great. As the
movement progresses the additional increment of load required to
produce a unit of diametral expansion grows progressively less till
a bulge of 0.052 is reached, after which there is a definite and con-
stant ratio between the increase of load and the expansion which
it produces. This ratio is 0.0065 for each increase of 1,000 pounds
in load.
It will be noted that in the case of the slate, just after the rock
began to deform, the curve shows a sudden break or sag which is
repeated at a second point before the regular movement, indicated
by the nearly straight line, is developed. This is due to the fact,
above mentioned, that the slate, being a foliated and not a granular
rock, is not isotropic in its response to pressure. It consists of little
INTERNAL FRICTION IN ROCKS 623
Pounds
100,000
90,000
80,000
70,000
60,000
50,000
40,000
30.000
20,000
10,000
-030 080 090 120 150 180 210 inches
Fic. 7.—True curve obtained by the deformation of a standard column of Carrara
marble in steel cylinder with wall 0.25 cm. thick. The area designated by oblique
lines represents the work done in effecting the deformation of the marble to a bulge
of 0.150 inch.
624 FRANK D. ADAMS AND J. AUSTEN BANCROFT
plates of kaolin and muscovite lying parallel to one another and at
right angles to the direction in which the pressure is exerted. The
breaking down of the foliated structure of the rock is indicated on
the curves by the irregularities to which reference has been made.
It will also be seen that in the case of granite, when the lateral
resistance is relatively low (e.g., when the rock is inclosed in the
steel tube having a o.25-centimeter wall), there is at the same
point a sag, though much less marked, due to the fact that the
lateral resistance offered by the tube is not quite sufficient to develop
a uniform movement in this the strongest of all the rocks employed
in the investigation.
Attention must be drawn to the manner in which deformation
goes forward in a column of rock when deformed under the con-
ditions of the experiment. As may be seen, if the tube and the
inclosed rock are sawed in two vertically, the column of rock begins
to move or flow at the middle, the motion taking place first along
the well-known shearing cones, having an angle of approximately
45° (usually somewhat greater), seen when a column or cube of the
rock is crushed between the faces of a testing machine in the ordinary
determinations of the strength of rock for building purposes. Thus,
as the movement progresses, there develops within the column two
obtuse cones, having as their bases the faces of the advancing pis-
tons and consisting of portions of the rock which show no evidences
whatsoever of deformation, but which are, under the conditions of
the experiment, subjected only to cubic compression. As the
experiment progresses, these cones (see A and B in Fig. 8) advance
into the deforming rock, additional amounts of the rock shearing
off the surfaces of the cones and thus coming to participate in the
movements which are going forward. Owing to the fact, therefore,
that the quantity of flowing rock is continually increasing in an
unknown ratio, it is impossible from the data mentioned above to
determine whether the definite increase in the ratio of load to
deformation is due to an increase of internal friction developed with
increase of pressure, or to the increased amount of material which is
being moved.
The answer to this question is obtained from another series of
experiments which exactly duplicated those with the columns of
INTERNAL FRICTION IN ROCKS 625
Carrara marble, described above, except that the lateral resistance
to movement was increased by increasing the thickness of the walls
of the steel tube inclosing the marble from a thickness of 0.25
centimeter to 0.33 centimeter. In these the amount of material
moved is identical with that in the series of experiments just
described, while the internal friction is increased by the increased
thickness of the steel tube.
A series of additional experiments were also made to determine
the resistance offered by such tubes when filled with soft tallow.
Fic. 8.—Longitudinal section through steel cylinder with pistons inserted and
inclosing a deformed column of rock—showing the obtuse shearing cones which advance
into the deforming rock.
In this way another series of curves were obtained for each
material and another “‘true curve”’ for the deformation of a stand-
ard column of Carrara marble under conditions identical with those
of the former experiments, except that the resistance to deforma-
tion offered by the steel tube was much greater. The “true curve”’
for the deformation of the marble in a steel tube having walls 0.33
centimeter thick is shown in Fig. to.
An inspection of this curve will show that while, as before,
starting from the limit of elastic expansion the rising load at first
induces a relatively small amount of movement in the rock, the
626 FRANK D. ADAMS AND J. AUSTEN BANCROFT
Pounds
100,000 —
Bea ee
$0,000
80,000
70,000
60,000
Slate
Belgian Blac
Dolomite
pee
Coes
50,000
\
\\
5 eee “ = -
f Alabaster a
\
40,000
30,000
(Copper)
Steatite
20,000
10,000
150 ‘210 inches
Fic. 9.—True curves obtained by the deformation of the several rocks when
inclosed in the steel cylinders with wall 0.25 cm. thick.
Pounds
100,000
90.000
70,000
60,000
30,000
20,000
INTERNAL FRICTION IN ROCKS 627
Diabase
Belgian Blac
Dolomite
Marble
Sandstone
Alabaster
—_—
—
— (Copper)
“210 inches
Fic. 1o.—True curves obtained by the deformation of the several rocks when
inclosed in the steel cylinders with wall 0.33 cm. thick.
628 FRANK D. ADAMS AND J. AUSTEN BANCROFT
ratio of the amount of this movement to increment of loads
increases rather rapidly, and, after deformation amounting to
about 0.06 has been brought about—which requires a load of
38,750 pounds—the ratio of increase of load to amount of deforma-
tion of the column becomes constant, as when the marble is
deformed in the tubes with thinner walls. It will be seen, however,
that for the experiments in the thicker-walled tube this ratio of
increase is much less than when the wall was thinner, i.e., 0.25
centimeter being 0.0051 diametrical increase for each increase of
1,000 pounds in the load, instead of .0065, as in the first series of
experiments.
This demonstrates that the moving rock possesses internal
friction and that with the increase of the lateral resistance the
amount or coefficient of friction rapidly increases, and at a con-
stant ratio.
The investigation was then extended to the other rocks of the
series enumerated on pp. 598 and 599. ‘The conditions and method
of conducting the experiments were in every case identical with
those just described with Carrara marble. Two sets of standard
steel tubes, having wall thicknesses of 0.25 centimeter and 0.33
centimeter, respectively, were employed, and the true curves were
plotted representing the mean of a series of experiments in each
case (see Figs. 9 and 10).
“WORK DONE” IN THE DEFORMATION OF ROCKS
If Px be the load to which the specimen is subjected and Py
be the resistance to movement offered by the inclosing walls of the
steel cylinder, the data were first examined to ascertain whether
the formula
Px—Py=a constant
represents the movement, and it was found that this was not the
case. ‘They were then studied to see whether each rock possessed
a constant factor K, which might be termed its modulus of plasticity,
as in the formula
Px—KPy=a constant
It was found that, if the data are calculated so as to take into
consideration the bulge of the cylinder and are plotted to show
INTERNAL FRICTION IN ROCKS 629
vertical stress as compared with lateral stress, this formula repre-
sents the facts and that each of the softer rocks possesses a definite
modulus of plasticity, this being also true in the harder rocks in the
earlier stages of the deformation at least.
This interesting fact is discussed at length in the accompanying
paper by Dr. King, where a mathematical treatment of some of the
new data developed in the present investigation is also presented,
illuminating certain parts at least of that hitherto unsubdued and
almost unoccupied domain—the mathematics of the flow of solids.
In the present paper, without entering into a mathematical
treatment of the subject, the following deductions from the experi-
mental data may be indicated.
If a vertical line be drawn cutting off the “‘true curve” obtained
in the case of any rock when the deformation of the tube amounting
to o.15 has been reached, and if the area inclosed by this line, the
‘“‘true curve’’ itself, and the base line of the diagram be measured,
this area represents the ‘‘work done”’ to effect the deformation of
the rock. This area showing the “work done” in deforming a |
standard column of Carrara marble in a 0. 25-centimeter steel tube
in Fig. 7 is shaded. In Fig. 9 the “true curves”’ obtained in this
deformation of all the rocks of the series, in steel tubes having a
wall thickness of 0.25 centimeter, are shown, and in Fig. 1o the
complete series of ‘‘true curves’? obtained when the wall thickness
of the tube is increased to 0.33 centimeter is set forth. In both
figures the curves are cut off at the ordinate 0.15, and the area
representing the “‘work done”’ in the case of each rock is clearly
shown and may be compared. .
Table III sets forth these comparative values in square inches.
This table shows quite clearly that with the increased resistance,
offered by the thicker-walled steel tube, the amount of work
required to effect an equal deformation increased in the case of
every rock. It also sets forth the comparative’ value of these
increases and also the relative amount of work done to deform the
different rocks of the series.
The table thus shows that the “work done” in deforming a
column of marble of the size employed and under the conditions of
the experiment, when inclosed in the thinner-walled tube, is to the
‘“work done” when an identical column is deformed, when inclosed
630 FRANK D. ADAMS AND J. AUSTEN BANCROFT
in the thicker-walled tube, as 51,708 is to 60,415. Or, again, that
the ‘‘work done”’ in deforming a marble column, whether the resist-
ance be small or great, is almost exactly one-half of that required
to effect an equal amount of deformation in a column of granite
under the same conditions. That is to say, almost exactly twice
as much work is required to deform granite as is required to effect
an equal deformation in the case of marble and nearly four times
as much as is required to produce an equal deformation in the case
of steatite.
TABLE III
RELATIVE AMOUNT OF ‘‘WoRK DONE” IN EFFECTING AN
EquaL DEFORMATION IN UNIT COLUMNS OF
DIFFERENT Rocks
UNDER RESISTANCE OF
o.25 cm. Steel Tube | 0.33 cm. Steel Tube
Steatites essen 26,054 34,123
‘Alabaster eke ssa 35,509 42,046
Sandstone: ).2)5: 1: 41,262 53440
Miarbletin cs jae). 51,708 60,415
Dolomite. 2). ....... 66,362 77,092
Belgian Black...... 735754 79,362
Slate weer aceon 79,009 07,154
Diabasenas asc i 92,985 107,431
Granites yaaa 104,169 | 119,877
TABLE IV
RELATIVE AMOUNT OF ‘‘WorK DONE” IN EFFECTING AN
EquaL DEFORMATION IN UNIT COLUMNS OF DIFFER-
ENT Rocks CALCULATED ON THE BASIS OF MARBLE
AS UNITY
UNDER RESISTANCE OF
o.25 cm. Steel Tube | 0.33 cm. Steel Tube
Steatitet sc . Gah. nee 0.50 0.56
Nlabastensse: tec. 0.69 0.71
Sandstone......... 0.80 0.88
Marblewess sie I.00 I.00
Dolomiten: ieee 1.28 1.28
Belgian Black...... 1.43 Tot
Slateta eee race TSS 1.61
Diabase css css fale 1.80 1.78
Granites aye ee 2.01 1.98
If the “work done” to deform marble be taken as unity, these
figures may be set forth as in Table IV.
INTERNAL FRICTION IN ROCKS 631
In these tables there is expressed in actual values the phenomena
which are displayed in such a striking manner in the great exposures
of the Grenville series and in other terranes which have undergone
deformation at great depths below the surface of the earth where
the same force has acted on a complex of rocks of diverse character.
In these occurrences some of these rocks are torn to fragments,
which are then carried far apart in a flowing matrix formed of some
other and more plastic member of the complex. This is seen in a
striking manner where dykes of diabase or belts of granite cut
through a limestone, and the whole complex is then deformed under
conditions of deep-seated differential pressure. The diabase dyke
or belt of granite is torn apart into angular fragments, which are
floated along in sinuous curves in the plastic flowing limestone, like
logs or drifting timber on the surface of a flowing river (see Fig. 11).
EFFECT OF A CHANGE IN THE RAPIDITY OF THE APPLICATION OF
PRESSURE
In Fig. 12 there are two curves: one showing the deforma-
tion of alabaster, the other, the deformation of marble. These
also illustrate the effects of a change in the rate at which the
pressure is applied.
In the former case, after a load of 36,000 pounds had been
gradually applied in successive increments and no movement had
taken place under the load for 2 minutes, the next increment of
load was by mistake applied suddenly, thereby submitting the
rock to an impact instead of to a slow increase of pressure. This,
as will be seen, produced at once a movement of 0.045 inch.
Following this, however, four increments of load, each of 1,000
pounds, had to be applied.before the movement was resumed, and
two additional increments, each of 1,000 pounds, had to be applied
before the movement could be re-established in its regular course,
after which the flow continued in the line followed by the normal
curve.
In the second case—that of the marble—the normal course of
the experiment was interrupted four times by postponing the time
of reading the deformation produced by a new increment of load
much longer than usual, namely, from 9 to 75 minutes. These
were when the load on the column of rock was 40,000, 55,000,
632 FRANK D. ADAMS AND J. AUSTEN BANCROFT
)
60,000, and 65,000 pounds, respectively. It will be noted that the
same effect, though on a smaller scale, was produced as that just
described as the result of impact. An abnormal increase of load
Fic. 11.—Photograph of a specimen of Trenton limestone which has been cut by
a narrow dyke of camptonite. The whole has then been distorted by pressure exerted
by the intrusion of the igneous mass constituting Mount Royal. The harder camp-
tonite has been broken into fragments which have been carried apart in the flowing
mass of more plastic limestone (Canadian Northern Railway Tunnel through Mount
Royal, Montreal, Canada).
INTERNAL FRICTION IN ROCKS 633
Pounds.
100,000
90,000
60,000
70,000
60,000
50,000
fe roees
40,000
30,000
930 “060 030 120 150 180 ‘Z10 Inches
Fic. 12.—Curves showing the effect of change in rate of application of pressure
634 FRANK D. ADAMS AND J. AUSTEN BANCROFT
was required to bring about a re-establishment of the movement,
which, however, eventually resumed its former course.
BEARING OF THE RESULTS ON CERTAIN PROBLEMS PRESENTED BY
THE EARTH’S CRUST
The experimental results afford a reply to the question pro-
pounded by Dr. Gilbert and set forth in the opening paragraph of
this paper. They also have a direct bearing on the problems pre-
sented by the origin of ‘‘decken” and by the theory of isostasy.
When movement producing deformation is once started in the
rock under the influence of tangential thrust, resulting in the break-
ing down of its texture, the rock, if deeply buried in the earth’s
crust, does not on that account offer a decreased resistance to
further movement.
Some experiments by Karman’ on the deformation of marble
under differential pressure have yielded data with reference to the
amount of this pressure which must be exerted in the case of marble
in order to induce plastic flow in the rock. The data obtained
represent maximum results, because in the experiments the pressure
was applied rapidly as compared with that which would be devel-
oped in any earth movements, and, also, the factor of heat was not
taken into account. It must be noted, however, that heat and a
very slow application of the deforming force would produce move-
ments under lower pressures than those made use of in the experi-
mental work. Karman found that, if a column of marble were
submitted to a supporting or containing pressure, such as that
exerted by the steel tube in our experiments, amounting to 685
atmospheres—which would be equivalent to that exerted by the
overlying strata at a depth of 2.53 miles below the surface*—it
would flow uniformly and continuously under a load of 2,870
atmospheres applied to the ends of the column. If the containing
pressure fell below the value mentioned, that is, if the rock occupied
a position in the earth’s crust nearer the surface, it would speedily
crumble and break to pieces, presenting in this way a failure similar
t “Festigkeits Versuche unter allseitigem Druck,” Zeit. des Ver. deut. Ingenieure,
October 21, IoIt.
2F. D. Adams, “Depth of the Zone of Flow in the Earth’s Crust,” Journal of
Geology, February, 1912.
INTERNAL FRICTION IN ROCKS 635
to that which is obtained in testing building stones in the laboratory.
On the other hand, if the containing or supporting pressure is
increased, the load required to produce deformation rapidly
increases also, and the experiments seem to indicate that with a
containing pressure of about 10,000 atmospheres, which would be
equivalent to a depth of about 22 miles below the surface, it would
be impossible to make the marble flow, except under a pressure
which would be simply colossal.
Since with the increase of resistance to tangential thrust, that
is, with increasing depth below the surface of the earth, the amount
of such thrust required to produce movements in the earth’s crust
increases rapidly, it is evident that the great movements of adjust-
ment by rock flow or transference of material in the earth’s crust
from one point to another—other than the transference of rock in
a molten condition—must take place comparatively near the sur-
face. That is, beneath the zone of fracture where adjustment takes
place by faults and overthrusts—in the zone of flow—movements so
far as they are determined by pressure are effected with an ease
which increases rapidly in proportion to their nearness to the
surface.
It would seem, therefore, that it is in the upper part of the zone
of flow only that the great ‘“‘decken,” as, for instance, those which
are developed in the Alps, are produced. ‘This explains the fact
that in the mountain range in question it is the upper “‘decken”’
which have moved more rapidly and have extended farther than
the lower “‘decken,”’ where the rock is under the increased load and
is consequently much less plastic.
Since with the increase of depth there is a rapid increase in
rigidity of the rocks of the earth’s crust, it is not difficult to under-
stand how it is that, while great movements may take place near
the surface of the earth in the upper part of the zone of flow, the
globe itself is ‘‘more rigid than steel or glass.”’
The experimental work also affords at least a first approxima-
tion to the determination of the dimensions of the forces which are
required in order:to effect deformation in the earth’s crust in the
case at least of the chief types of rocks which make up the crust
in question.
636 FRANK D. ADAMS AND J. AUSTEN BANCROFT
In these measurements it must again be noted that the factor
of pressure alone was considered, no account being taken of the
element of heat in the crust, which would undoubtedly tend to
increase the ease of movement.
In the experiments it has been shown, as mentioned, that the
resistance to deformation exerted by the wall of the steel tube
gradually increases as the experiment progresses. If, however, the
value of the resistance is taken at a point where the regular column
shows a diametral increase of 0.05 inch (or 6.35 per cent), i.e., when
the deformation is well under way and after which it becomes pro-
portional to the increased tangential pressure, this resistance, in the
case of the experiment with the steel wall o.25 centimeter thick,
would be equivalent to 26,685 pounds to the square inch, or 1,815
atmospheres, that is, to a depth of 4.2 miles below the surface.
In the case of our experiment with a steel wall 0.33 centimeter
thick it would be equivalent to 37,359 pounds per square inch, or
2,542 atmospheres, that is, to a depth of 5.8 miles below the surface.
Thus at these respective depths the additional tangential thrust
required to induce a pronounced movement in the case of marble
and granite, respectively, would be as shown in Table V.
TABLE V
At DrptH OF 4.2 MILES At Deptu oF 5.8 MILes
Pounds pet. Square Atmospheres Pounds "en Square Atmospheres
Marbles taste. 66,400 4,517 74,500 5,008
Granite eee. 138,500 0,422 159,000 10,857
CONCLUSIONS
t. All the rocks employed in the present investigation can be
deformed under differential pressure at ordinary temperatures.
2. In order to effect an equal deformation, it is necessary to
employ differential pressures having different values in the case of
the several rocks.
3. The ease with which these rocks are deformed has as one of
its functions the hardness of the rock (or of the minerals compos-
ing it).
INTERNAL FRICTION IN ROCKS 637
4. In the case of the softer rocks—alabaster, steatite, marble,
etc.—the deformation is produced by movements due to a slipping
within the constituent crystals of the rock on their gliding planes,
often accompanied by twinning, the movement in this case being
similar to that seen in metals when they are deformed. In the
harder rocks the deformation is accompanied by granulation, the
texture developed being similar to that found in mylonite.
5. Each of the softer rocks at least has a well-defined modulus
of plasticity.
6. The “‘work done”’ when a rock is deformed by a tangential
thrust, within the earth’s crust, increases rapidly with the weight
of the superincumbent strata, i.e., with its depth below the surface.
7. The relative ease with which the several rocks will flow under
differential pressure is shown in Tables III and IV, which give
mathematical expression of the ‘“‘ work done”’ in deforming standard
columns of each rock.
8. A uniform thrust exerted ona prism of the earth’s crust may
deform and fold the upper portion of the mass, while it will be quite
insufficient to produce any movement in the lower part of the same
mass.
9. The thrust required to develop deformation, taking no cog-
nizance of the influence of heat or the time effect which might result
if the pressure were applied with extreme slowness, in the case of
marble, and of granite, is shown by the values given in Table V.
10. To revert to the question propounded by Dr. Gilbert, in
order to develop flow in any rock within the earth’s crust the rock
must be submitted to a differential stress which is greater than that
which is required merely to break down its texture and very much
greater than that which is sufficient to crush it to pieces under the
ordinary conditions which obtain at the surface of the earth.
ON THE MATHEMATICAL THEORY OF THE INTERNAL
FRICTION AND LIMITING STRENGTH OF ROCKS
UNDER CONDITIONS OF STRESS EXISTING IN THE
INTERIOR OF THE EARTH
LOUIS VESSOT KING
McGill University, Montreal
INTRODUCTION
That solid bodies could be permanently deformed and made to
flow without rupture under sufficiently great stress has long been
known. ‘The extensive experiments of Tresca on the flow of metals"
(1864-72) directed the attention of several mathematicians of the
time to the subject. Tresca announced as a result of his experi-
ments the simple law that a stressed solid would commence to flow
as soon as the maximum shearing stress exceeded a limiting value K
characteristic of the solid. This hypothesis was incorporated into
the elastic solid theory by Saint-Venant? and others. The hope
was expressed by these writers that by effecting the solution of
simple problems in ‘“‘plasticodynamics,’’ corresponding to the
experimental arrangements employed, it might be possible, not
only to verify the theoretical results, but also to determine a specific
constant K characteristic of the various metals and related in an
intimate manner to other physical constants. It was found pos-
sible, however, to solve only a very limited number of extremely
simple problems: (1) circular cylinder under uniform pressure over
the plane ends or subject to uniform lateral pressure; (2) cylindrical
shell constrained to remain of constant length and subject to uni-
form internal and external pressure; (3) circular cylinder twisted
beyond the elastic limit; (4) bar of rectangular section bent by a
suitable distribution of forces to take the form of a circular arc.
*H. Tresca, Par. Mém. Sav. Etr., XX (1872), 75 ff. and 281 ff. A summary of
Tresca’s experiments is given by L. S. Ware, Journal of the Franklin Institute, LX-XIII
(1877), 418 f.
2 Saint-Venant, Comptes Rendus, LXVII (1868), 131 ff., 203 ff., 278 ff.; LXVIII
(1869), 221 ff., 2900 ff. ’
638
FRICTION AND LIMITING STRENGTH OF ROCKS 639
None of these simple problems corresponded, however, to any
detailed observations available. The position with regard to the
final mathematical interpretation of Tresca’s observations was
summed up by Saint-Venant in a communication to the French
Academy.* It was stated that, before much progress could be
made in formulating a mathematical theory of plastic flow, it would
be necessary to plan experiments more easily capable of mathe-
matical specifications; in particular he recommended that means
be taken to trace out in the znierior of the solid the extent of the
plastic deformations. The difficulty of doing this without at the
same time interfering with the continuity of the solid under test has
apparently not been overcome up to the present, so that data on
plastic deformation available for mathematical treatment are still
very meager.
It is interesting to notice, however, that we have available at
the present day a method of exploring the internal structure of
solids which seems to fulfil the need expressed by Saint-Venant.
By the use of extremely powerful X-rays it has been found possible
to detect internal cavities in steel castings not visible on the sur-
face. The subject has recently been extensively studied by Davey,’
who states that it is possible to detect an air-inclusion 0.021 inch
thick in 1} inches of steel and an air-inclusion 0.007 inch thick in
3 inch of steel. More recently Pilon,’ making use of the Coolidge
tube, has successfully penetrated 5.5 centimeters of steel. This
method appears to the writer to offer the means of studying in
successive stages the plastic deformation of specimens of various
materials under conditions of intense stress. In these circumstances
it would be necessary only to drill extremely fine holes in the
specimen in various directions and to study the deformation of
these as the solid is made to flow.
Tresca’s hypothesis that flow in a solid commences and continues
as long as the shearing stress exceeds a definite limit has been found
t Saint-Venant, ‘‘De la suite qu’il serait nécessaire de donner aux recherches
expérimentales de plasticodynamique,” Comptes Rendus, LXXXI (juillet, 1875),
II5—21.
2.W. P. Davey, Trans. Am. Electrochem. Soc., XXVIII (1915), 407-18.
3H. Pilon, Rev. de Mét., XII (Nov., 1915), 1017-23.
640 LOUIS VESSOT KING
by later tests to be only approximately true. It is found that to
produce continuous flow in a plastic solid it is necessary continuously
to increase the distorting stress. A simple illustration of this fact
is to be noticed in the manner in which a short circular cylinder
crushed in a testing machine ultimately breaks down. According
to Tresca’s theory the surfaces of shear should be cones of semi-
vertical angle of 45°, while experiments indicate that the angle is
more often in the neighborhood of 55° for a material like cast iron."
These results have led to a modification of Tresca’s hypothesis as
already mentioned. The effect of this so-called ‘‘resistance to
flow”’ does not appear to have been studied with a view to formu-
lating the laws according to which solids may be made to flow
continuously. .
In the field of experimental ballistics the use of the permanent
deformation of short copper cylinders to measure the enormously
high pressures involved in testing explosives by means of the
so-called ‘‘crusher-gauge,” invented by Noble about 1875,? has
led to the detailed study of the relation of applied stress and
deformation produced in these special circumstances. The results
of these observations have recently been studied in detail by Bril-
louin.t The behavior of copper shows the existence of internal
friction analagous to that observed by Adams and Bancroft in the
case of various rock specimens.
In the experiments carried out by the latter investigators the
use of nickel-steel jackets of standard thickness to incase the rock
specimens subjected to flow is analagous to the use of short cylinders
of annealed copper in the crusher-gauges just referred to. In order
to obtain the lateral pressure on the specimen corresponding to a
given deformation of the nickel-steel jacket, a calibration-curve is
obtained by filling the cylinders with tallow. The hydrostatic pres-
sures required to give a series of deformations give the required
tA. Morley, Strength of Materials (Longmans, Green, & Co., 1908), p. 55.
2 See Encyclopaedia Britannica, 11th ed., article on “‘ Ballistics,” for a brief descrip-
tion of the crusher-gauge.
3 Vieille, Mémorial des poudres et salpétres (Gauthier-Villars, Paris),V, 12-61.
4M. Brillouin, ‘‘Les grandes déformations du cuivre par écrasement et par
traction,” Ann. de Chimie et de Physique, 9° série, II (1914), 489-96.
FRICTION AND LIMITING STRENGTH OF ROCKS 641
calibration-curve, Just as the copper cylinders of the crusher-gauge
are calibrated under known end pressures in a testing machine.
MATHEMATICAL DISCUSSION OF THE OBSERVATIONS OF ADAMS AND
BANCROFT DURING THE ELASTIC STAGE
Although the experiments which form the subject of the present
discussion were all carried out when both rock and nickel-steel had
been deformed beyond the elastic limit, it is not without interest,
especially in view of further experiments on the subject, to follow
out the distribution of stresses in the rock specimen and in the
nickel-steel throughout the elastic stage. The necessary theory
from which the formulas given below are derived has been given by
the writer in a previous paper.’ Asin that discussion, it is sufficient
for the present purpose to consider the ideal problem of plane stress,
that is, one in which the end pressures and lateral pressures are such
that the displacements at the outer surfaces, both of the rock
specimen and of the nickel-steel jackets, are everywhere symmetrical
with respect to the axis and everywhere constant for a given load.
In reality the nickel-steel jacket shows a bulge over the center of
the specimen. As long as this is not too great the analysis will give
an approximate representation of the state of stress in the central
portion of the specimen and nickel-steel jacket at which the measure-
ments of displacement were taken by means of a sensitive exten-
someter. ‘The justification for this mode of treatment has already
been noticed in the writer’s paper previously referred to in its
application to a similar problem.
We denote by 7? the stress component along the radius 7; by 60,
the component at right angles to r; and by 22, that along the axis.
According to Lamé’s notation, w is the modulus of rigidity of the
rock specimen and 2d one of the moduli of elasticity such that
k=(A+3u) is the modulus of compression. Poisson’s ratio is
denoted by c=3A/(A+u). We denote by accented symbols the
corresponding elastic constants for nickel-steel. In the problem
under discussion we denote by 0 the radius of the rock specimen and
tL. V. King, “‘On the Limiting Strength of Rocks under Conditions of Stress
Existing in the Earth’s Interior,” Journal of Geology, XX (February-March, 1912),
T2520
642 LOUIS VESSOT KING
the interior radius of the nickel-steel jacket, and by c the exterior
radius of the nickel-steel jacket. If P is the pressure per unit area
applied to the end of the test specimen, we have 22=—P. The
principal shearing stresses are one-half the algebraic difference of
the principal stresses and are at once obtained by writing a=o in
the equations (13) of the writer’s paper mentioned above. We then
obtain
() $|#—2|=3— py B
lire Os o I+B
(ii) 3|7#%-68|=0 (1)
Ae re Mid 1—20¢, £B
(ii) | —m |= pp} 74 Ft
where
ee Eo : I
Bee ne Ue aueg! CG 1—b?/¢? (2)
The radial displacement U at the outer surface of the rock specimen
is given by |
UWP. o B
pine 2 “t+o1+f
(3)
Each of the principal shearing stresses (i), (ii), (ili), is associated
with a family of surfaces along which the material will crack or
flow. These are illustrated in Fig. 1, reproduced from the writer’s
paper already mentioned. It is important to notice in the present
connection that the principal shearing stresses in the interior of
the rock, as given by (i) and (iii), are independent of the radius r
and remain equal throughout the elastic régime. It thus follows
from Tresca’s theory that the rock, when stressed under these ideal
conditions, will commence to break down or flow szmultaneously
throughout its entire volume. The surfaces of shear which will be
associated with the elastic breakdown may either be the system
of cones (i) of semivertical angle 45° or the system of helicoidal
surfaces (iii) of 45° pitch giving rise to the well-known Luder’s lines
on the curved surface of the specimen. The particular surfaces of
shear which will be observed in any particular test will depend on
accidental circumstances, as either system is equally likely to occur.
FRICTION AND LIMITING STRENGTH OF ROCKS 643
We easily derive expressions for the principal shearing stresses
in the nickel-steel jacket. At points distant 7’ from the axis these
are
SN: al, poe pee (on : I c?/r'2—t
OS a ee i Mg Od iat cs/ Oot
+7 => a; Cc I CA) r?
f 3 '— 96 =
(ii) 2 | rr | Jeane 1+, c2/b?—1 (4)
AD 2/2
Gi) alma -ap
1—o0 1+ B 2?/b?—1
The radial displacement U’ at the outer surface of the nickel-stee
jacket is given by
UR I ge
cm @ 1—8/e 1460! (5)
By writing u=o and therefore B=o, c=3, the foregoing give the
familiar results for stresses in a cylinder subject to internal hydro-
Sy
4@
ne
Aa
Ts
rc
\
\\
x
"
<>
Q
C2
“f
|
\\\\
eee
()
2
8
~
}:
S82
22.
\ \\
?
%
6
$36
—
~
Sas oe
$2
©
oe
x
x
x
oO
I
\
%,
<5
oN
/
,
“
see
\
505
|
Hh
|
\
yD
egees
\
\ i
SES
Xr
040
eee
XK
)
5
\
\\\
SS 052
SeS
\
x)
x)
x
Soot
00,
OX?
sent
|
\
oe
|
\\\
|
(ii) ( iii)
Fic. 1.—(Figure from this Journal, Vol. XX, No. 2 [February-March, 10912],
p. 123.)
static pressure. The three principal shearing stresses given above
all take their maximum value (independent of sign) at the interior
644 LOUIS VESSOT KING
surface r=b, and of these maxima (ii)’ is the greatest. The maxi-
mum shearing stress is therefore
Bi Ohman ag? eae (6)
It follows from this discussion that elastic breakdown of the
nickel-steel jacket commences at the interior surface and, as defor-
mation continues, extends gradually to the outer surface. The
surfaces of shear in this case are the system of cylindrical surfaces
whose traces on a plane perpendicular to the axis of the cylinder are
equiangular spirals intersecting orthogonally and cutting all radii
at angles of 45°. An examination of the nickel-steel jackets shows,
in fact, that the surfaces of shear approximated roughly to this
system. The polished outer surface of stressed specimens showed
indications of fine longitudinal ribs, while in such as were actually
ruptured it was noticed that the surface of rupture conformed to
that predicted from theory. As the rupture occurred when the
nickel-steel was stressed very much beyond the elastic limit, the
actual surfaces of shear are determined by very complex conditions
involving the effect of internal friction, with which we shall deal in
a later section.
Numerical results —A rough verification of the preceding results
may be made by calculating the relation between the load and the
increase of diameter of the nickel-steel jacket according to equa-
tion (s). For nickel-steel we take o’=0.327 and w’=10.8X 10°
pounds per sq. in., values employed in the writer’s paper just
referred to. In one set of experiments (referred to as o. 25-centi-
meter wall) b=1.00 cm., c=1.25 cm., giving from (2)
be
B al aas 68X Pane i mi .
When the jacket is filled with tallow we may take o= 3, u=0, B=0,
so that equation (5) gives
U’/c=1.34X (P/e’),
or in terms of the total load, W=7b?P, we obtain
2U’ (inches) =2.52X1077XW (pounds) (7)
FRICTION AND LIMITING STRENGTH OF ROCKS 645
When a specimen of Carrara marble is inserted we have’ c=0. 2744,
M=3.154X10° pounds per sq. in., whence 6=1.889 and
U'/c=0.176X(P/p’), or 2U’ (inches) =3.31X10~?XW (pounds) (8)
In another set of experiments (referred to as 0.33-centimeter
wall), b=1.00 cm., c=1.33 cm., giving
It+o pw
I—o p*
B=2.96X
In the case of tallow filling we find as before,
U’/c=0.980X(P/p’) or 2U’ (inches) =1.96X10—7XW (pounds) (9)
and in the case of the Carrara marble specimen
U'/c=0.147X(P/p’) or 2U’ (inches) =2.94X107-°XW (pounds) (10)
In Fig. 2 are compared the observed and theoretical stress-strain
diagrams corresponding to the cases calculated out in equations
(8) to (10). In the case of tallow filling, the initial slope of the
observed curves agrees approximately with the calculated slope.
In the case of the marble filling, the agreement is within the limits
of error involved in measuring these extremely small strains.
MATHEMATICAL DISCUSSION OF THE OBSERVATIONS OF ADAMS AND
BANCROFT DURING THE PLASTIC STAGE
1. Navier’s theory of internal friction.—Let xx, Vy, and 22 be the
principal stresses in the solid at a point P measured toward the
origin (Fig. 3). Let S be the shearing stress in a plane whose
direction cosines with respect to the direction of the three principal
stresses are (J, m, n). Let N be the stress normal to this plane.
We then have
SHN = Pee my hee |
and (11)
N=Pxx+myy+n722
Generalizing somewhat on Navier’s hypothesis of elastic break-
down, we may state that the material will not break down as long as
SG (12)
t Adams and Coker, ‘‘Elastic Constants of Rocks,” Publication No. 46 of the
Carnegie Institution of Washington, 1906, p. 69.
646 LOUIS VESSOT KING
where K is a function, not only of the stress V normal to the plan
at which slide occurs, but also of the previous history of the
Bulge
W. load in
fo) 2 4 6 8 ro ue a 16 18 2¢ thousand lbs.
Fic. 2.—Theoretical and observed stress-strain diagrams. Curves 1, tallow
filling. Curves 2, Carrara marble filling.
specimen. According to Tresca’s hypothesis, K was regarded as a
constant, depending only on the nature of the specimen. An exten-
FRICTION AND LIMITING STRENGTH OF ROCKS 647
sion of this hypothesis due to Navier (the so-called internal-friction
theory) replaces (12) by the condition
S<K+phN,
mu being a new constant somewhat analogous to the coefficient of
friction of mechanics. In order to
discover the relation between the
principal stresses at the elastic limit,
it is necessary to find the direction
(1, m, n) which makes (S—yuN) a
maximum and equate the result to
K. Suppose the principal stresses to
be all of the same sign, two of
them equal, yy=%%, and 22>x2
(corresponding to the state of
affairs in the cylindrical rock speci-
mens under test). We then have, Hee
writing /=sin @ cos ¢, m= sin 6 sin ¢, n=cos 8,
S24 N?= 4%? sin? 06+-22 cos? 9, N =a sin? +22 cos’O | oS
S = (%—£z) sin 6 cos 6 ce
S—pN = (—X2) sin 9 cos 6—p(xx sin 8 +22 cos? 8) (14)
This expression reaches a maximum when
cot 20=—p, (15)
in which circumstances
(S—pN) max. = 3 (22 cot 6—£x tan 9), (16)
and the relation between the principal stresses at breakdown is
given by
s3=2K tan 6+ xx tan? 0 (17)
where @ is given in terms of u (the coefficient of friction) by (15).
This result indicates that the material in question will break down
along a family of cones of semivertical angle a=37—0.
2. Discussion of observations.—In the experiments of Adams and
Bancroft the cylindrical rock specimens were subjected to end loads
transmitted by the steel pistons. Asa result of the intense pressure
648 LOUIS VESSOT KING
developed, the rock cylinders were caused to bulge out laterally
over the central portion, where the thickness of the nickel-steel
jacket was reduced to 0.25 centimeter and 0.33 centimeter, respec-
tively, in the two sets of experiments. The rock was thus sub-
jected to a continuous succession of breakdowns, so that it was
possible from these observations to
determine the relation between the end
and lateral pressures required to keep
the rock in movement.
Considering the central portion of
the rock cylinder throughout which the
flow takes place, we may reasonably
assume, when the bulge is small, that
the average pressure-intensity P, along
the direction of the axis is given by
Po=Wo/ (x05),
W, being the load on the steel piston and .
b, its radius. As the bulge becomes
sensible, it is necessary to make a cor-
rection to allow for the increasing area
over which the pressure is distributed.
Referring to Fig. 4, we denote by P the
average pressure-intensity across a plane
at right angles to the axis at the position
Fic. 4 of maximum bulge where the radius of
the cross section is b. We denote by p
the resultant traction per unit area exerted by the nickel-steel jacket
on the rock specimen in a direction making an angle e with the axis.
Then, considering the equilibrium of one-half of the rock specimen,
we may write
tP s+ [p cose dS=7b*P, (18)
the integral representing the total component of the tractions
between the rock specimen and the nickel-steel jacket in a direction
parallel to the axis of the cylinder. When an exactly similar jacket
is filled with tallow and deformed by the application of a load on
the steel pistons in the same way, we may’ consider the pressure in
FRICTION AND LIMITING STRENGTH OF ROCKS 649
the interior to be hydrostatic. If p, be the hydrostatic pressure
required to bulge ‘the nickel-steel jacket to the same radius 0, we
have instead of (18) the equation
TPobe+ [Po COS &odS = 7b? po, (19)
where €, now denotes the direction which the normal to the deformed
surface makes with the axis of the cylinder. It was carefully ascer-
tained in the experiments of Adams and Bancroft that the shape of
the bulged nickel-steel jacket was the same when occupied by the
softer rocks and such an easily flowing metal as lead, in which con-
ditions of pressure approach very nearly to hydrostatic conditions
under the very intense loads employed. As the deformation of the
nickel-steel jacket is due to the distribution of surface tractions ),
it is reasonable to assume that they are distributed in approximately
the same way. This is equivalent to asserting that the tangential
component of the surface traction between rock and nickel-steel is
negligible compared to the normal component, a statement which
seems to be reasonable in view of the fact that both rock and nickel-
steel are highly polished over the surface of contact. We may thus
write [p cos e dS=Jp, cos €, dS in (18) and (19) and arrive at the
relation
P= p.(1—63/b?) + Pob3/0", (20)
giving the average pressure-intensity at the center of the specimen
to be identified with 23 of equation (17). The corresponding lateral
pressure is given by ~, which is identified with £% of (17).
We are now in a position to test the theory of internal friction
expressed by (17) from the observations of Adams and Bancroft.
It is only necessary to plot against each other the end pressures 22
and the lateral pressures ¥% as determined above. Such specimens
as give straight lines may be said to possess a definite modulus of
plasticity, K, and coefficient of internal friction, u. Curves obtained
in this way are shown in the Appendix, where they are described
in detail for the various specimens tested. The results show that
for some kinds of rock the curves approximate closely to straight
lines between certain limits of pressure. In the interpretation of
these curves it must be kept in mind that the material is not broken
\
650 LOUIS VESSOT KING
down from an initially unstrained state" at each stage of the process.
The constants of plasticity and internal friction, as determined by
the present investigation, refer to rock which is being made to flow
continuously. This state of affairs, however, approaches more
nearly to that occurring in nature during slow geological deforma-
tions than to conditions existing when the rock is broken down from
an initially unstrained state.
Under ideal conditions the curves for the observations taken
with the nickel-steel jackets of the two wall thicknesses should be
identical. Actually, however, they differ to some extent, indicating
that the effect of stresses set up by the deformation of the nickel-
steel has not been entirely eliminated. The two sets of observa-
tions are, however, sufficiently close to give approximate estimates
of the relation between the principal stresses which must exist
before the rock can be made to flow under conditions existing in
the earth’s crust. It will be noticed from the curves of Plate I that
for the harder rocks, such as diabase and granite, the curves along
which breakdown takes place show the existence of a very large
coefficient of internal friction. Since the hydrostatic pressure is
given by 4(2x%-+22), this is equivalent to the statement that the
stiffness or limiting shearing stress required to break down the
rock increases with the hydrostatic pressure to which the rock is
submitted. In other words, we come to the important conclusion
that the stress-difference required to break down rock material under
conditions of pressure existing im the earth’s crust increases with the
depth. In the application of this result to geophysical problems,
the foregoing conclusion may have to be somewhat modified to
take into account the rise of temperature with depth. It is highly
desirable that further experiments be carried out with a view to
ascertaining the influence of this factor.
NOTE ON APPLICATIONS TO GEOPHYSICAL PROBLEMS
Up to the present the only quantitative data available for use
in geodynamical problems have been obtained by crushing cubes
of various rocks in a testing machine according to the ordinary rules
Compare Karman’s observations on marble and sandstone, Zeit. des Vereins
deutscher Ingenieure, October 21, Ig1t.
FRICTION AND LIMITING STRENGTH OF ROCKS 651
of engineering practice. The unsatisfactory nature of such data
as applied to conditions of stress deep down in the earth’s crust has
already been pointed out by the writer... The results now available
from the observations of Adams and Bancroft supply much needed
data for the purposes of geophysics. Quoting from a classical paper
by Sir George Darwin,’ ‘‘With regard to the earth we require to
know what is the limiting stress-difference under which a material
takes permanent set or begins to flow rather than the stress-
difference under which it breaks; for if the materials of the earth
were to begin to flow, the continents would sink down, and the sea
bottoms rise up.” In the paper quoted Darwin estimates roughly
the stress-difference in the interior of the earth due to a distribution
of continental masses corresponding roughly to the actual distribu-
tion. For instance, it is estimated that the stress-difference under
the continents of Africa and America is at a maximum at more than
1,100 miles from the earth’s surface and amounts to about 4 tons
per square inch. Darwin’s conclusion that ‘‘marble would break
under this stress, but that strong granite would stand”’ must be
modified considerably in the light of the results of Adams and
Bancroft, as the limiting strength of the rock material under the
enormous pressure at the depth referred to would probably be
increased many times. For the purposes of such calculations the
curves of Plate I may be employed as they stand. If, for instance,
it is desired to investigate the stability of mountain ranges or of
continental elevations, the principal stresses at great depths must
be derived from the theory of elasticity, making use of elastic con-
stants derived from the interpretation of seismological records. If
the principal stresses at any point be plotted as 22 and £% on such
a diagram as that of Plate I, a particular rock material will flow
if the point falls between the axis 22 and the curve characteristic
of the particular rock formation under consideration. The material
will be on the point of flowing if the point falls on the curve itself,
while the rock will stand the stress if the point falls between the
tL. V. King, oP. cit., p. 120.
2 Sir G. Darwin, ‘“‘On the Stresses Caused in the Interior of the Earth by the
Weight of Continents and Mountains,” Phil. Trans., CLX XIII (1882), 187-230;
Scientific Papers, IL (1908), 405.
652 LOUIS VESSOT KING
PLATE I
zz longitudinal
pressure
thousand
lbs. per
sq. in.
180
160
Granite
140
Diabase
120 Slate
Dolomite
Carrar: 1
Fan a Marble
Sandstone
80 Alabaster
Steatite
60
Lead
Tallow
xx lateral
pressure
thousand lbs.
per sq. in.
RELATIVE PLASTICITIES OF VARIOUS ROCK SPECIMENS
Nickel-steel jackets of 0.25 cm. wall
Belgian Black Marble
FRICTION AND LIMITING STRENGTH OF ROCKS 653
EO Cl
zz longitudinal
pressure
PLATE II
thousand
Ibs. per
| sq. in.
180
| Granite
160 Diabase
Slate
140 poe Belgian Black Marble
Dolomite
120 Carrara Marble
Sandstone
a Alabaster
Steatite
| Lead
Tallow
t xx lateral pressure
thousand lbs.
per sq. in.
RELATIVE PLASTICITIES OF VARIOUS ROCK SPECIMENS
Nickel-steel jackets of 0.33 cm. wall
654 LOUIS VESSOT KING
curve and the axis %. Thus for complete stability the entire series
of points representing stress-differences beneath a continental eleva-
tion must fall in this latter region. -It is thus evident that the
existing theories of isostasy should, in considering the equilibrium
of stresses called into existence by continental elevations and moun-
tain ranges, take account of a “‘compensation of plasticity’’—1.e., of
the increased stiffness or resistance to deformation—of the under-
lying rock when submitted to greater hydrostatic pressure. With
the reservation already made as to the possible influence of tem-
perature, we have a considerable basis of evidence in favor of the
conclusion that at any time in the past history of the earth continen-
tal elevations might have attained much greater altitudes above sea-
level than any at present existing, without giving rise to stress-
differences in the earth’s interior sufficiently great to have caused
rupture or breakdown, owing to the much increased “‘resistance to
flow’’ set up in the rock material by the great pressure of the over-
lying crust. We should conclude also that, in the event of flow
occurring, the region of flow would be confined to a region of the
earth’s crust comparatively near the earth’s surface. The increas-
ing limiting stress, with pressure characteristic of rock material
made to flow as in Adams’ and Bancroft’s experiments, leads one
to the conclusion that great movements of the earth’s crust have
for the most part always proceeded by extremely slow and con-
tinuous adjustments to pressure conditions, and not, as supposed
by some geologists,’ by a series of cataclysmal collapses of the type
which would occur if the material of the earth’s crust possessed in
all circumstances a unique and definite limiting strength analogous
to that obtained by crushing a specimen, unsupported laterally, in
a testing machine. The further consideration of these problems
must, however, be left over for further discussion. Enough has
been said to make it evident that the results of Adams and Bancroft
have provided much needed data in the light of which many of the
existing theories of geodynamics may require considerable modi-
fication.
1G. A. J. Cole, Presidential address delivered before the geological section of the
British Association, Manchester meeting, 1915, B. A. Report, pp. 403-20.
FRICTION AND LIMITING STRENGTH OF ROCKS 655
APPENDIX
In order to study the experimental data on the flow of rocks in the light
of a theory of internal friction, the data reproduced in Tables I and II were
obtained from the original large-scale curves obtained by Adams and Bancroft
connecting the end load on the steel pistons with the bulge of the nickel-steel
jacket. Each of the curves represented the mean of two, three, or more com-
plete sets of observations. The first row of numbers for each specimen is the
total load W. (in thousand pounds) on the steel pistons required to bulge
the nickel-steel jacket by the amounts entered under the various columns.
The second row gives the pressure-intensity P,= W 7b; in thousand pounds per
square inch exerted on the end of the specimen of radius }). The third line
gives the average pressure-intensity P=22 in thousand pounds per square inch
at the central portion of the specimen in the direction of the axis, correcting
for the effect of the bulge from formula (20). It will be noticed that the
average longitudinal pressure at the center is somewhat less than that over
the ends by amounts which increase considerably with the harder rocks.
The final results given in Tables I and II are shown graphically in Plates I
and II, respectively. Against the lateral pressures (given by the experiment
on tallow) are plotted the longitudinal pressures required to bulge the nickel-
steel jacket to the same extent. For such of the rocks as give curves approxi-
mating to straight lines we may say that a definite modulus of plasticity and
coefficient of internal friction exist. Rough estimates of these constants as
determined for the soft rocks from large-scale curves are given in Table III,
in which the first entry corresponds to the o.25-centimeter wall nickel-steel
jacket and the second to the 0. 33-centimeter wall. It will be noticed that the
two sets of results are in poor agreement for K, and are only in rough agreement
for p, the difficulty arising from attempting to fit a straight line to a series of
points which are only approximately colinear.
In the case of the harder rocks no definite coefficient of friction can be said
to exist. In the case of dolomite and Belgian black marble it is noticed that
the coefficient of friction tends to diminish with increasing longitudinal and
lateral stress. Slate gives a very irregular curve due to the development of
cracks while the material is stressed. The sudden bend in the curves for
diabase and granite is attributed to the actual breakdown of the rock material.
From the curves of Plates I and II it will be noticed that this occurs in the
neighborhood of 22=150,000, *%= 25,000, corresponding to a stress-difference
of 125,000 pounds per square inch. In a general way this result confirms the
conclusion already arrived at from a discussion of the experiments of Adams
on the pressure required to close up small cylindrical cavities in specimens of
Westerly granite. In the writer’s paper already mentioned (p. 641, n. 1) it
was pointed out that the stress-difference required to break down the rock
material in the neighborhood of small cavities amounted to as much as
osna
aao
nh
OOH
HH
AonMm!/ aot! +tHwW!] Hot! oma
to 0
nHeO] Mat
mow
non
HO
Ono
teh
LOUIS VESSOT KING
OHN
arn
mtr o
omo
TMM] OHM] OHO
onnm)] “t0
~omm
656
viz L°0z £61 L°g1 v'v1 2
tere g'oz Dt: L°Qt Fond : SOUUUCOOOUOOERGROSD ie
vor I‘O1 9°6 1°60 ol g
v-SSt o' rst : 6° Obr g:orr ORGS G Gee |OO eS OTs
L°9St 6 gSt g 1S O'ztr 6°911 eA IOLy, OR OO BOY OD GPO BH (niRtS)
VLE S-gL o'bL 0° 69 OWL Sipe eens: om
S'Lz1 €°ee1 es 9 gir 6° 601 “| 0:6 ROP ODE Ps
Sook Sher Besar Orr ©: +6 Freestar bless sense ee sens gseqerg
a Xe) L°09 SaiQs o'vs OO Vases | erences om
Z°QII (4°v11) | (6°11) (4°v1r) | (€° gor) vee
g‘ozt (L:o11) | (orn) | (4°S1x) | (£-gor) fries ccogsfeccc Fees estes ss segaptc
g's (6°98 ) | (4°98 ) | (vgs ) | (OES ) [roe
L°€o1 6° Io1 £° oor 9°40 OSE Qe ree nemaresie 22
“Sor L:£o1 $*Ior +96 5 tah 0 Fare fasten haa o shee
bis 90S S- 6+ op zed PEST apqieul YovIq urisjeg
9°16 £°6g £18 bill gid ON Obes
$66 9:00 z°99 o'gL gil [ec og }\rrttttt Foo ++ + -gqrm0foq.
S-Sv z' by o- ey o'ge ose Sey y |
9°z9Q 8°09 g 6°SS Sica o-St
9°£9 9°19 9S ges 1S Dette eeeeeee ee grazer
oe oof S:Lz z°Qz O°zz
S°os o-gs v'oS oss z°oS OO Ls
S-09 99s 6:9S bss £06 vere esog teense eee es “++ > guoyspueg
$°6z 1° gz L Le o lz S'bz CPOE ACY |
g zs LoS 6°9t I'gv 6°9¢ goeaseds 2 2
bes sro5 xy) €or vot ole SSK Ory, ED OOS EEO SCH SISO hip
o'9z 0° Sz ob Q°zz o°gI cee OY]
I'byv zey grb z°ov g°9f ‘eg
Let L-Sb Bop oie 6:9€ teres sorter tree reser eee es aq17e01S
giz £-1z S:o0z L°61 O°gI ROE
£61 o'O1 I'gi 6°91 ver LL = 22
L°61 o:' 61 I'gt 6°91 v'S1 Tee 0 chaimp hoes: asec oedate MOT] ery
9°6 z°6 88 zg $9 a
€zo'o 0z0'0 ¥IO'O OI0'0 ¥oo'o |-2¢/2g—1) ae
lol | ,rS6L- | ,,v86L- | ,rr6L: | ,roge: [eos Qe tile Sr Or ee
4, O10°O | ,, BOO"O | ,, gOO°O | ,, Foo'o | ,, zoo°o | °gz—qz na
TIVM AALANW
5 ¥ 2 g i -ILNA)-S$z°0
I WIaVve
FRICTION AND LIMITING STRENGTH OF ROCKS
Il Wave
Lev ov voc £°ge v'of pve SOLe: O° gz L' gz g'Sz $° vz S-0z es OY
Lev rv 9g 6e v'ge S°9f SUAS g ie I'gz L°9z L°Sz g bz $:0z poncnaemOrt
fmexd 0°0% £61 £°g1 g 41 8°91 vst L°€1 o'e1 S21 o'7r o'or aay, |
ZLL1 I'gli 9° gt 6'gLr z°QLI OLLI volt z°Idy Igor I'fg1 g rst (Giese ZK) in) | Viera sa:
g £12 g Loz 6° 002 €°L61 z°eO1 1'6g1 6° gr g bli o'rlt z° Sor z° OSI (LSE CiT) ey | ea ory,
£- yor z* 101 g 40 0°96 o'6 o°z0 $°6g 1°Sg ze z°O8 ool (0°09 ) “OW
9'vOL g FoI L°¥g1 6° tor L°tg1 8 vor z°zgl o'zSt S°Lb1 I'ovi £°gz1 g iL
L°LO1 I‘ 101 L°Vgi v-Igt z°QLt o'SL1 6°g9o1 I°SSr o'OS1 girl vole o'ril
z°96 0° £6 6°68 £°9g8 L°98 o'Sg zB S°SZ o'eh 0°69 o°z9 oss
£:oSt L-oSt S-oStr o°ZSI S:zS1 C2EST L°1St (1°zS1) | (¥°zSr) I'zSt v'ori 9g ez1
g 6OL1 vvlt v° gor 8°991 g ror 9° zg1 O° LS1 (V2SSz)— |) (o7SSz) o'vS1 Livi neck dp Wok eee od
Vlg 4°+g g'1g g°0g z°0g 0°64 L494 (Sesh )= | (8° SL-) o'sh 0°69 0°09 IY |
S°PeI £°2e1 I'6zI €° Ler g ver = I'2Z1 Q gir L°81r fi eit z Sor L°go1 o' SO eo eee 74
$-6S1 o'zSt g err o°Of1 I'VEI 0° 621 €°Lz1 O°S11 z° Vir Q'III g Lor e250 : OF
ghd Ibs ool g 49 v'S9 6°z9 1°09 S308 Ass Sigvsi etal S-ov Pa Oy)
veer I'Of1 O°92I ever L121 V' gir €°vr1 L°go1 o'ror €°oOo1 y 9° $6 6°Sg T fees 44
6°9S1 z OVI o'ovI g S€1 QOfl o'Ser S°gi1 £° gor g' Sor S*I01 v°96 TQ Q% —e) | Meas oF
S-of Lael £°g9 1°99 8 £9 0°19 g 4s oes Sar S'6P oly o-2v ey, |
S-g1I vert $*goI I‘ for 6°36 1'+6 £°98 L°SL 6°oL v'g9 z°£9 zeS tees egg)
L°ge1 6° Ler €°L1r 4°11t g Sor 8°36 z°6g omrey4 girl 1°69 9° £9 SaeS EEO
£°99 v'z9g z2-Ls Svs S'1s zt S- ey SOAS o'se £:¢ ore 0°92 SOOT
v- Lor L°fo1 0° g6 z°S6 3°06 £°9g bog o vl fees ris z°39 v°99 aay
oO Sz €°L11 gL°o1 Zor 996 b-00 o'£g z°Sh eh gis L°39 9°99 pas
o-19 zLs S°2s I'o$ 1-Ly 1'vv S:ov L°9¢ L°S¢ ose see S°2e ae
z° £6 0°16 +99 £98 9° £3 g 62 g vZ 9°49 v'So v°z9 o-gs 1°6v ECE
o' Lor 6101 v°96 1°26 9°88 beg O°oL L°g9 z°99 o'£9 ¢-9s z°6V BOS
z-eS L°6v oly z'Sy z'ey Lov S-L¢ S288 £°ze SL°of S° gz o' bz GeO |
vg g'1g LLL I°SZ S12 I°l9 L°19 g'Ss 4°eS 61S £°6v 6°gt CE OOF
6°SO 1°06 o'tg 0°08 z’ss 1°69 £°£g vgs £°vs £°zS 9 6t o°6£ OO IOS 2
89 z by o'r o'6¢ Lg o've 6°0f S'Lz S'9z $°Sz Z'vz o'61 CERO
I'ev gov v'gt vLe Sse (Sats £:0f L'gz 6°Sz 5 O'vz vez 6-02 £0 = 22
1'tv gor v'ge PLE Oks 88 £:0f L°9z 6°Sz 6 bz vez 6°02 esa =.
S-0z g°O1 £°9g1 e°Qi flr z°Qt L’vi ofl gz1 I°zt bri zor TSAO AA
£1z°o0 gL1I'o gt1°o SIr‘o S$60°0 Ilo'o groo £zo'°o 0z0'oO vio'o o10'oO ¥oo0'o |*2q/2q—1
uVL83° VL93° iy Vlyg: Wp VLeQ" WVlZeg’ VLIQ° VLOg° yVLOL: VSL i VEOL: VIOL: FLAY Ga || R0 08-0 gz
4, COLO | ,, OBO'O | ,, OJO'O | ,, OSO'O | ,, OVO'O | ,, O£O'O | ,, OZO'O | ,, CIO" | ,, gOO'O | ,, gOO"O | ,, VOO'O | ,, zoO'O | ~Gz—gz
ZI II OI 6 8 4 9 $ v £ z I
tees ses peor
ayers)
hss asequiq
testes see gqerg
a[qieul Yorpq urispg_
“+++ aqTuT0T0q
“os aiqaeyy
“++ Quoyspues
Dieses eeesees suaqseqery
"8+ aqrye04g
“sss money,
wVLgl o=qz
aang
TIVM YAILIN
-ILNAD-££°0
658 LOUIS VESSOT KING
160,000-200,000 pounds per square inch. The conclusion, there limited to
small cavities, is extended by the present experiments of Adams and Bancroft
to continuous rock stressed under conditions approaching those existing in
TABLE III
Specimen K va
(Pounds per square
inch
Steatites ashe. 5,500-1,800 0. 24-0.32
Alabastenssacrie- « 4,200-3,100 0.37-0.38
Sandstone....... 7,500-3,100 ©.34-0.40
Marble ®ajeaclese 850-1,500 ©.58-0.52
Weed eee OR ae 850- 500 ©.00
the earth’s interior, in which circumstances a limiting stress-difference
several times greater than that obtained by the usual crushing test must
be employed.
NOTE ON THE DEPOSITS CONTAINING HUMAN RE-
MAINS AND ARTIFACTS AT VERO, FLORIDA
E. H. SELLARDS
Geological Survey of Florida, Tallahassee, Florida
The deposits containing human remains and artifacts at Vero,
Florida, have been described in the issue of the Journal of Geology
of January-February, 1917.7 Among interpretations advanced in
connection with that discussion, that proposed by Dr. R. T. Cham-
berlin differed in one important respect from that offered by
the writer. To Dr. Chamberlin it seemed that the fossils found
in the stream bed, or most of them, were secondary, having been
washed in from a near-by older formation, while the writer held
that these fossils were primary. In order, if possible, to harmonize
these views, Dr. Chamberlin very considerately returned to Florida
for the purpose of re-examining the deposits and was present with
the writer at Vero from March 16 to March 20, 1917. Professor
E. W. Berry, who is engaged in a study of the fossil plants, was
also present, as well as Mr. H. Gunter and Mr. Isaac M. Weills.
The additional collections made include potsherds, bone imple-
ments, flints, vertebrate and plant fossils.
The banks of both the main and the lateral canals were re-
examined, and it was found that the upland formation from which
Dr. Chamberlin assumed that the vertebrate fossils had washed was
almost if not entirely non-fossiliferous. These new observations
have served to define more closely the problems to be solved.
It is evident that the fossils found at this locality are primary
fossils in the stream bed and were not washed in from the older
formation of the uplands near by. It is quite obvious also, both
on old and on new evidence, and in conformity with views
previously expressed, that the human remains and artifacts of this
deposit do not represent burials by human agency as was main-
tained by Dr. Hrdlicka in the former discussion. ‘The strati-
graphic evidence on this point is so conclusive that it is certain
that the hypothesis of burial by human agency may be eliminated
« “Symposium on the Age and Relations of the Fossil Human Remains Found at
Vero, Florida,’’ XXV (1917), 1-62.
659
660 FE. H. SELLARDS
as a possibility. It thus appears that the problem is confined to a
study of the stream fill, and that the determination of the age of
the human remains depends upon a correct understanding of the
history of accumulation of material within the stream bed.
The view which the writer holds regarding the stream deposit
has been expressed in papers previously published. The earliest
phase of deposition in this stream bed includes local accumulations
of muck which fill holes and channels in the underlying marine
shell marl. Another phase which is general and is observed through-
out most of the stream valley includes light-colored, often cross-
bedded, sands which pass at a higher level into brown or dark-colored
sand. This part of the deposit has been designated as stratum
No. 2. Above this brown sand is found as a rule an accumula-
tion of alternating layers of sand and muck, stratum No. 3, which
when fully developed is capped by a fresh-water marl. The
maximum thickness of the stream fill as preserved at the present
time is from 4 to 8 feet. The first human bones obtained were in
the brown sand beneath the fresh-water marl, and additional
bones were subsequently found in this sand. A flint spawl was
found in place in the sand and additional flints and two bone imple-
ments were obtained from siftings. From the alternating layers
of sand and muck which lie above the brown sand human bones
and artifacts have been taken in considerable numbers.
As already stated, human bones and implements have been
taken from beneath the fresh-water marl. The last bone imple-
ment collected on the recent visit to the locality (Florida Survey
collection No. 7786) was found beneath this marl and’ lay at a
depth of 4 feet from the surface. The place of this implement is
on the south bank 32 feet west of the lateral inlet canal. The bank °
at this place is relatively high and has retained its capping of
fresh-water marl. It is evident, therefore, that the human materials
of this deposit were not accumulated by the recent stream. On the
contrary, they lie in deposits accumulated at an earlier stage.
The writer’s interpretation, as expressed in papers previously
published, is that the human remains and artifacts are contempo-
raneous with the extinct vertebrates of this deposit, and that the
age of the formation, according to the accepted Neen of
faunas and floras, is Pleistocene.
THE FOSSIL PLANTS FROM VERO, FLORIDA
EDWARD W. BERRY
Johns Hopkins University, Baltimore, Maryland
The discovery of human remains associated with an extinct
mammalian fauna at Vero, Florida, has excited a great deal of local
and general interest, and various theories regarding the age of these
remains and the manner of their occurrence have already been
advanced, and admirable accounts of the local geology have been
given by Sellards and others. It is therefore unnecessary for me
to repeat any of these details in connection with the present pre-
liminary abstract of my study of the fossil plants.
Plant remains in the form of laminae of impure peat or scattered
fruits, chiefly acorns, are present from the bottom to the top of the
deposits overlying the shell marl which forms the base of the section.
The lower sands (designated No. 2 by Sellards) have yielded no
leaves and but few acorns, but the upper bed (Sellards, No. 3) con-
tains many leaf layers alternating with sand laminae, and it is from
the latter horizon that all of the plants enumerated in the following
pages have been collected, with the exception of one species of
acorn which is common to both beds.
Recent and extinct mammalian and other bones occur in both
layers, and human remains are also found in both beds. After a
thorough study of the local sections and the paleontologic evidence
.l am convinced that there is no hiatus between beds Nos. 2 and 3
and that there is no great difference in age from the bottom to the
top of the section, although it records changing physical conditions
and necessarily becomes gradually more and more recent as the top
of the section is approached. The lower sand marks the recession
of the sea in which the underlying shell marl was formed. The
tE. H. Sellards, Am. Jour. Sci. (IV), XLII (1916), 1-18; Eighth Ann. Rept.
Florida Geol. Surv., 1916, pp. 122-60, Pls. 15-31; Science, N.S., XLIV (1916), 615-17;
Jour. Geol., XXV (1917), 4-24, Figs. 1-4; R. T. Chamberlin, ibid., XXV (1917),
25-39, Figs. 1-9; T. W. Vaughan, ibid., pp. 40-42; A. Hrdlitka, ibid., pp. 43-51,
Figs. 1, 2; O. P. Hay, zbzd., pp. 52-55; G. G. MacCurdy, ibid., pp. 56-62, Figs. 1-6.
661
662 EDWARD W. BERRY
upper beds (No. 3) mark successive seasonal layers of valley filling
in the narrow valley of a small stream. This stream was apparently
always small, and the marvelous abundance of fossils at this one
point seems to be due to a bar or sinkhole or similar cache formed
near the junction of the two lateral branches which united near this
point to form the main stream. The determinable plants are
represented almost exclusively by fruits or seeds, as the leaves, with
the exception of the coriaceous oaks, which are abundant, were too
thoroughly decayed before they were buried to retain their identity.
The study of such remains is beset with many difficulties. The
material has to be sorted without allowing it to dry. It then has
to be impregnated with paraffin simultaneous with drying. Finally,
identification is hampered by the lack of recent material for com-
parison, and when the material is identified the determination of the
exact range of the still existing species on which so much hinges
is a matter of great uncertainty in the present state of our knowledge
of plant geography. Iam under obligations to, and take this oppor-
tunity of thanking, Mr. W. L. McAtee, of the Biological Survey, for
determining five species of fruits for me.
The following plant species have been identified from the Vero
deposits:
Pinus taeda Linné
Pinus caribaea Morelet
Pinus sp.
Taxodium distichum (Linné) Rich.
Carex sp.
Pistia spathulata Michx.
Seronoa serrulata (Michx.) Hooker
Sabal palmetto (Walt.) R. & S.
Myrica cerifera Linné
Leitneria floridana Chapman (?)
Quercus virginiana Mill.
Quercus Laurifolia Michx.
Quercus Chapmani Sargent ( ?)
Quercus brevifolia (Lam.) Sargent
Polygonum sp.
Magnolia virginiana Linné
Anona glabra Linné
Brasenia purpurea (Michx.) Caspary
Ilex glabra (Linné) A. Gray
Acer rubrum Linné
Zizyphus sp. (new species)
Vitis cf. rotundifolia Michx.
Vitis sp.
Benzoin cf. melissaefolium (Walt.)
Nees
Viburnum nudum Linné
Viburnum cf. dentatum Linné
Xanthium sp.
The most abundant form in the preceding list is Quercus lauri-
folia, which is represented in the upper beds by leaves, cupules, and
acorns and in the lower beds by cupules and acorns.
It is still
FOSSIL PLANTS FROM VERO, FLORIDA 663
found growing in the Vero region, but is not nearly so abundant
as far south as Vero at the present time as it appears to have been
at the time the Vero deposits were laid down.
SUMMARY
The foregoing comprise more or less positively identified remains
of twenty-seven species of plants. Nineteen of these have not
been previously found fossil, while the following eight have already
been discovered in Pliocene or Pleistocene deposits:
Pinus taeda Brasenia purpurea
Taxodium distichum Acer rubrum
Quercus virginiana Zizyphus sp.
Magnolia virginiana Viburnum nudum
The problem in so far as it relates to the evidence of the plants
depends on the correct evaluation of the change which this plant
assemblage shows when compared with the flora now growing at
Vero.
Of the plants found fossil the following are still found at Vero,
and I have included in this list as probably found in the present
flora of Vero the four forms of Pinus, Carex, Polygonum, and
Xanthium which are not specifically identified. This list comprises:
Pinus caribaea Quercus brevifolia
Pinus sp. Polygonum sp.
Carex sp. Magnolia virginiana
Serenoa serrulata Ilex glabra
Sabal palmetto Acer rubrum
Myrica cerifera Vitis cf. rotundifolia
Quercus virginiana Xanthium sp.
Quercus laurifolia
In addition to the foregoing fifteen species still found at Vero
the following two species are found growing within ten or twelve
miles of Vero:
Taxodium distichum Anona glabra
Three species approach to within about fifty miles of Vero, being
recorded from the southern extension of the lake region flora of the
central peninsula in De Soto County. These are:
Pinus taeda Viburnum nudum
Pistia spathulata
664 EDWARD W. BERRY
The following six species are not now found growing in peninsular
Florida:
Leitneria floridana ( ?) Vitis sp.
Quercus chapmani (?) Benzoin cf. melissaefolium
Brasenia purpurea Viburnum cf. dentatum
Of these Leitneria floridana is a very local form not found nearer
than the Apalachicola River, and the chief center of growth of
Quercus chapmani is also in west Florida, while the true Viburnum
dentatum does not occur nearer than the upland region of
Georgia.
Finally, the Vero deposits have yielded a fruit probably identical
with similar remains from the late Pleistocene of New Jersey
representing an entirely extinct species of Zizyphus, a genus
abundant in Southeastern North America during the Tertiary,
but not now represented except by a single species of the arid south-
west (Texas to Arizona).
Two of the fossil species have been recorded from the Pliocene.
These are Taxodium distichum and Magnolia virginiana. One,
Quercus virginiana, is found in the early Pleistocene of both Ken-
tucky and Alabama, and the following occur in the late Pleistocene:
Pinus taeda Acer rubrum
Taxodium distichum Zizyphus sp.
Quercus virginiana Viburnum nudum
Brasenia purpurea
These latter, while they constitute but 26 per cent of the known
fossil flora at Vero, are especially significant in connection with the
fact that they all occur elsewhere in the physiographically youngest
of the Pleistocene terrace deposits, namely, the Talbot of New
Jersey and Maryland, the Chowan of North Carolina, and the
corresponding lowest terrace at several localities in Alabama, while
the Vero deposits constitute the youngest physiographic terrace
plain of the region and are referred to the Pensacola terrace by
Matson."
In my judgment and in the ordinary acceptance of that term
this flora is unquestionably of late Pleistocene age.
«G. C. Matson, U.S. Geol. Surv., Water Supply Paper 319, 1913, Pp. 31-35.
FOSSIL PLANTS FROM VERO, FLORIDA 665
Regarding its bearing on the interesting problem of the age of
the human and associated mammalian and other remains at Vero,
my study of the locality furnishes the following somewhat categori-
cal conclusions. ‘The underlying shell marl which forms a definite
and undisputed datum plane is late Pleistocene in age. Its species
all exist in near-by waters at the present time, and many of them
have been recorded from shell marls found from southern New
Jersey to the Florida keys and forming a part of the lowest and
latest well-defined terrace plain previously mentioned as having
been named Talbot in Maryland, Chowan in North Carolina, and
Pensacola in Florida. It follows that the vertebrate remains which
are so numerous at Vero cannot possibly be of Middle or early
Pleistocene age unless they are regarded as having been reworked
from older deposits, and I cannot conceive that this was possible,
nor do the vertebrate paleontologists who have examined the de-
posits consider that such was the case. In fact, I believe that, if
it had not been for the overestimate of the age of this vertebrate
fauna, Dr. Chamberlin would not have advanced his hypothesis
of the reworking and mechanical mixing of these bones and
Dr. Hrdlicka would not have insisted on the human burial theory
to account for the presence of the human skeletal remains. Nothing
is more reasonable than to suppose that the larger elements in the
Middle Pleistocene fauna of more northern areas should have
lingered for thousands of years in this more genial southern clime
until the presence of man in considerable numbers and the changing
climate as is attested by the fossil plants should have brought about
the extinction of a large percentage of the fauna. The fauna itself
confirms the rather limited data furnished by the fossil flora of this
change in climate, since it indicates a more mesophytic habitat
than exists today in the vicinity of Vero. Regarding the burial —
theory of Dr. Hrdlitka it may be said that a part of the plant
material came from immediately above one of the human skeletons,
and I cannot conceive of the possibility of not being able to see
evidence of artificial burial in material made up of alternate layers
of sand and matted leaves and other vegetable débris. I therefore
see no reason to doubt that relative modern men were contempo-
raneous with this partially extinct fauna of Middle Pleistocene.
666 EDWARD W. BERRY
aspect which survived in Florida to the late Pleistocene. With
regard to the exact age of the Vero deposits there are, it seems to
me, but two alternatives, and these apply equally and are in large
part derived from a study of the physiography and the faunas and
floras of the corresponding topographic forms. in the other states
of the Coastal Plain. These alternatives are that they are about
the same age as the Peorian interglacial deposits of the Mississippi
Valley or are immediately post-Wisconsin and correspond with
what the Scandinavian geologists have named Litorina time.
FURTHER STUDIES AT VERO, FLORIDA f-
ROLLIN T. CHAMBERLIN
University of Chicago
In a recent number of this Journal there appeared a symposium
on the age and relations of fossil human remains found near Vero, ©
Florida.t The several investigators attacked the problem from
quite different viewpoints and developed considerable difference of
opinion. Later Dr. Sellards planned to spend an additional week
at Vero with Dr. E. W. Berry in further study of the critical points
involved in the case. He was good enough to invite the writer to
join in this further inquiry, and this invitation was cordially
accepted. As a result of the wider examination of essential points
made possible by my second visit, I desire to amend and extend
the interpretation of the history of the human bones and associated
relics previously offered.
My studies on the first visit were given almost wholly to an
endeavor to work out the physical history and time relations of
the formations at Vero, as this was regarded as a step necessary to
the safe interpretation of the relics embraced in them. This
seemed especially necessary because the dates of the appearance of
man and of the disappearance of the extinct animals were among the
very points brought into question and could not themselves be used
as decisive criteria. On the other hand, the nature and successions
of the formations afford some of the most critical evidence bearing
on these dates. It will perhaps be recalled that the history of the
formations was found to be rather definitely deployed, and that
the time relations of the deposits were quite well indicated by the
physical criteria available, irrespective of their fossil contents.
My former reading of this history was confirmed in all essential
1. H. Sellards, R. T. Chamberlin, T. W. Vaughan, Ale’ Hrdlitka, O. P. Hay,
and G. G. MacCurdy, ‘‘Symposium on the Age and Relations of the Fossil Human
Remains Found at Vero, Florida,” Jour. Geol., XXV (1917), 1-62.
667
668 ROLLIN T. CHAMBERLIN
particulars by what was seen during the second visit. Its essen-
tials are here recalled for the sake of the discussion following.
t. During a submergence of this portion of the east coast of
Florida there was laid down a striking marine shell marl which has
sometimes been called “coquina.” It is the oldest formation
exposed to view and has been referred without question to the
Pleistocene. Though its precise place within the Pleistocene has
not been determined, its fauna was essentially the same as that
now living in the adjacent ocean. Following the deposition of the
marine shell marl, a withdrawal of the sea gradually brought this
region into the horizon of terrestrial action. In the transition,
beach conditions prevailed, resulting in sandy deposits, partly
marine, partly terrestrial.
2. At the appropriate stage in the withdrawal of the sea a
barrier ridge was developed immediately to the west of the present
location of the Florida East Coast Railway. This ridge parallels
the railroad and the coast for many miles both north and south of
Vero, and throughout most of its extent it is a pronounced topo-
graphic feature. West of it was a marshy area.
3. With further withdrawal of the sea a newer barrier ridge
developed from two to two and one-half miles east of the earlier
Vero beach ridge. This constitutes the present east coast of
Florida. For over one hundred miles it incloses, between itself
and the mainland, a salt-water lagoon, known as the Indian
River.
4. After the withdrawal of the sea from the Vero beach ridge,
erosion developed a channel in essentially the position now occupied
by Van Valkenburg’s Creek. The very low gradient and notable
width of this channel in proportion to its very insignificant depth,
which was limited by sea-level, suggest that erosion, which here was
slow at the best, was in progress for a considerable time.
5. In the marshy region west of the Vero beach ridge bog
deposits accumulated here and there. Cementation had also
affected certain horizons of the sands of this tract and had con-
verted them into a sandstone. This had been effected by the depo-
sition of iron and manganese oxides as well as organic matter in
the sands. The length of time involved in this process of conver-
FURTHER STUDIES AT VERO, FLORIDA 669
sion of the sands into sandstone may well have been considerable,
though it cannot be definitely measured.
6. But, whatever the length of this period, it is important to
observe that the filling of the channel of the creek did not begin until
after the sand had been converted into black sandstone, for water-worn
pebbles of this black sandstone are abundant in the basal portion of
Fic. 1.—The present channel of Van Valkenburg’s Creek, dry since the con-
struction of the drainage canal in 1913. Shows the relatively slight depth of the
channel.
the channel fill. ‘They are in fact rather more conspicuous at the
base of the fill than at higher levels, although occurring throughout.
The special significance of these black pebbles, as brought out in
the symposium, lies in the fact that they fix the date of the filling
of the channel with respect to the old bog area to the west. The
oldest fill in the creek channel is notably younger than the bog
deposits of the uplands back of the main beach ridge.
7. The filling of the creek channel from this beginning up to the
present has taken place in two stages, which appear to be quite
670 ROLLIN T. CHAMBERLIN
distinct in some portions of the channel, but which at some other
points can be separated only with much doubt. At best they are
thin, both of them together averaging only 5—7 feet in thickness, and
they are quite changeable in composition. ‘The lower of these has
been designated formation No. 2 by Sellards and the upper one
formation No. 3. In this paper the former will be termed the lower
creek deposit and the latter the upper creek deposit. The bones
and relics in question were found in these two creek deposits.
The discrimination of these successive stages of formation made
it seem quite possible that the land life of the times began to occupy
this region during the stages of emergence, and hence that bones of
the extinct mammals and other vertebrates might have accumulated
in the marshy area to the west of the Vero beach ridge in Pleistocene
times, following not long after the coquina stage, and that later,
as Van Valkenburg’s Creek gradually cut back into this area, these
old Pleistocene bones were washed into the stream channel and
concentrated in the creek deposit, while at this later time there
mingled with them relics of the more recent vertebrates and plants,
as well as human remains.!. Thus the deposit of the stream channel
might contain fossils of quite different ages in intimate association.
The geologic conditions and the sequence of events seemed such
as to suggest and to support this hypothesis.
On the assumption that the extinct mammals were perhaps as
old as Middle Pleistocene—as was then urged—and that the
coquina formation which underlies the region could not well be
interpreted as much older than this—if indeed as old, since all of
its fossils belong to living species—there seemed to be rather urgent
reasons for presuming that at least the older of the extinct mammals
invaded the region as soon after its emergence from the sea as con-
ditions permitted. They were therefore supposed to have been
present during the formation of the marsh deposit back of the
beach ridge, and to have, in all probability, been buried in it, and
their relics derived from it in the subsequent trenching and filling
by Van Valkenburg’s Creek. The finding of balls of black sand-
stone from the marsh deposit in both the older and the younger
creek deposits seemed to fit at once, and help explain, this very
t Symposium, pp. 25-39.
FURTHER STUDIES AT VERO, FLORIDA 671
puzzling combination of bones of extinct animals of supposedly
Middle Pleistocene age mingled with fragments of human pottery
of almost obvious recency.
The actual presence of bones of the extinct animals in this
Pleistocene marshy area was not observed, for, on the first visit,
time did not permit an adequate examination. And so a leading
Fic. 2.—The formations of the creek section exposed by digging into the south
bank of the canal near point marked WN in Fig. 4. No. 1 represents the marine shell
marl (coquina) grading upward into light-colored sands; No. 2 is the lower creek
filling of variously stained sands (Sellards’ formation No. 2); No. 3 represents the
upper creek filling (Sellards’ formation No. 3), consisting of alternate layers and lenses
of sand and muck; No. 4 is the loose dump material piled on the surface in excavating
the canal.
purpose of the second visit was to search the older upland forma-
tions for direct fossil evidence on this point. This search was not
successful in finding bones of the extinct animals, either iz situ, or
in the canal dump from the upland area through which the two
forks of the creek have cut. The conditions that prevailed at the
time of the marsh deposit to the west of the barrier ridge seem to
have been inhospitable to life of the types of the extinct animals
672 ROLLIN T. CHAMBERLIN
in question, or else the nature of the formation was unfavorable to
their preservation. This, of course, is not conclusive evidence that
they did not then live in the region, but it greatly weakens the
hypothesis that bones deposited in these beds were sources of
supply to the creek deposit after the analogy of the black pebbles.
Bones as well as coarser fragments of any durable material
should, of course, tend to become concentrated in a stream bed as
the finer inclosing sands are washed downstream. ‘This is a well-
recognized principle and it might well account for the fact that
bone fragments are rare in the upland formations and numerous in
the creek channel deposits. But whether this selective concentra-
tion of coarser fragments in the channel by the action of the stream
is quantitatively adequate to explain the difference is questionable,
and it is not wise to appeal to it unless all other explanations fail.
The solution of the riddle of the mixture of the bones of extinct
animals with human bones and pottery was therefore sought on
other lines. uy
It is true that, at a point three miles west of Vero, Dr. Sellards
had found the wreck of a proboscidian in a fresh-water marl deposit
close to the surface and referable to the general upland deposit
back of the beach ridge.’ Dr. Sellards had also recognized a fauna
similar to that found in his formation No. 2—the lower creek
deposit—in a fresh-water marl bed belonging to the upland deposit
at a point about 1,700 feet east of the Florida East Coast Railway
bridge, i.e., downstream from the deposits which contain human
relics. Both these facts seem to imply that a fauna of the general
type found in the lower creek deposit occupied the region at some
time during the formation of the upland deposits, and to this
extent they support the general correctness of the inferences enter-
tained in my contribution to the symposium, but they do not sup-
port the specific view that the bones of the lower creek deposit were
in any large measure derived from the lagoon, or marsh, deposit
of which the indurated black sand is a part.
These facts also weaken the presumption that the relics of the
extinct animals really imply so great age as Middle Pleistocene.
Dr. Hay, who favored the view that they were closely related to
t Symposium, p. 55.
FURTHER STUDIES AT VERO, FLORIDA 673
the fauna of the Aftonian inter-glacial beds of Iowa, yet recognized
that ‘this fauna might have continued on for another stage or two,
but by the time of the Illinoian drift it had become essentially
modified.”* It is further to be recognized as not improbable that
this fauna may have lingered longer at the south than it did at the
north, where the advances and retreats of the ice border were
putting the fauna under the stress of an oscillating climate.
The marine coquina deposit, which lies below all the upland beds
and the creek deposits as well, does not bear evidence of great age,
its shells being all of living species. This deposit, or perhaps more
strictly the beach sands into which it grades upward, are referred
by the geologists of this and adjoining states to what has been
termed the third or lowest Pleistocene marine terrace formation.
The age of this terrace was assigned by Matson to late Pleistocene.”
There are good reasons, therefore, in the stratigraphy and the
topographical aspects of the deposits at Vero, for regarding the
extinct mammals and other vertebrates as continuing to a relatively
late date. The aspects of the problem thus developed made a
closer scrutiny of the two creek deposits more imperative, for, as
we have seen, both of these deposits were late in the history of the
formations of the region, and the oldest of these formations bears
both a paleontological and a topographical aspect of relative
recency.
This closer scrutiny at the time of the second visit developed
evidence both for and against the point previously made by
Dr. Sellards that the delicate condition of the fossils—as well as
their grouping—was not consistent with the view that they were
derived from an older formation by stream action. Dr. Sellards
put forward an increasing number of fossil remains which, on
account of their fragile nature, or because of the close association
of various bones, he did not believe could have suffered transpor-
tation or much disturbance since fossilization. That an occasional
specimen of this sort need not be of much significance was pretty
effectually established by the finding, among a half-dozen fragile
1 Symposium, pp. 54-55.
2Tbid., p. 40; G. E. Matson, ‘“‘Geology and Ground Waters of Florida,’
U.S. Geol. Surv., Water Supply Paper 319, 1913, pp. 31-35.
674 ROLLIN T. CHAMBERLIN
carapaces of turtles, of one specimen still firmly holding together
and undoubtedly still capable of being swept along by a stream for
a considerable distance without being torn to pieces. But the
cumulative evidence of the cases presented suggested strongly
that various particular individuals, at least, were primary fossils
but little disturbed since entombment.
On the whole, it seems to me that Dr. Sellards has strengthened
his view that at least an essential part of the fossils of the lower
creek deposit are primary to that deposit, and that the extinct
animals represented by these fossils were denizens of the region
as late as the formation of the lower creek deposit, Sellards’ forma-
tion No. 2. This does not entirely dispose of the hypothesis that
some of them were washed in from the older deposits in the process
of stream-cutting and stream-filling, but it renders that possibility
less vital to the essential question—the time of man’s appearance
in this region. At the same time, of course, it brings the time of
extinction of the fauna of the lower creek deposit down to a rela-
tively recent date.
It, however, left the critical feature of the problem—the admix-
ture of extinct animals with human remains, pottery, and bone
implements of modern aspect—still crying for a satisfactory expla-
nation. The crux of the whole problem seemed to be thrown upon
the trustworthiness of the discrimination between the upper and
the lower creek deposits. Now these upper and lower deposits
altogether measure only about 6 feet in thickness on the average.
This is a pretty thin deposit to divide into two distinct ages when
the natural irregularity of such deposits is considered, and when the
composition of the earlier and the later deposits is so nearly the
same as it is in this case. The upper creek deposition reaches
down to the year 1913, when the digging of the drainage canal put
an end to the activity of this portion of Van Valkenburg’s Creek,
and thus the occurrence within it of pottery, bone implements, and
the remains of man is altogether what one might expect. But the
presence of these same relics in the lower creek deposit would tell
a different story. It therefore becomes imperative to note sharply
in just what portions of the creek filling the significant relics were
found. It is also equally important to determine critically in what
FURTHER STUDIES AT VERO, FLORIDA 675
horizons within the creek deposits the undisturbed individuals of
the- extinct vertebrates occur. Creek deposits, by their very
nature, imply changing conditions from time to time.
Dr. Sellards had appealed to, as evidence against the secondary
nature of the fossils of the old vertebrates, a number of bone
assemblages, such as a tapir skull, a wolf’s head, an armadillo,
Fic. 3.—View showing a distinct dividing line between the lower creek filling
(No. 2) and the upper creek filling (No. 3). Lenses and irregular patches of material
in both formations rapidly pinch out, showing considerable scour and fill. Location
close to that of Fig. 2, but at a different stage in the progress of the exploratory digging
in March, 1917. At the base is the underlying formation No. 1; at the top is the
canal dump piled on the surface. Note the thinness of the creek fillings.
turtle carapaces, etc., which he did not believe could have been
moved since fossilization. In going over the list one by one with
Dr. Sellards, it developed that the fossils of old extinct forms which
seemed to him to necessitate the belief that they have not been
rewashed, were found in formation No. 2 (the lower creek deposit)
and in general rather well down in it. Here must perhaps be
676 ROLLIN T. CHAMBERLIN
excepted the turtles, but the finding of one very firm carapace near
the junction of the two deposits would seem to throw much doubt
upon arguments based upon the turtles. If it be admitted, then,
that such of these fossils as cannot readily have been transported
from elsewhere since fossilization are primary to the lower creek
deposit, that would mean that this earlier creek filling is of the
same age as these particular types, and so its age is determinable
from these types provided they afford decisive evidence. But a
development of scarcely less significance in the ultimate interpre-
tation was the bringing out of this very fact that the undisturbed
specimens of extinct vertebrates were taken wholly, or chiefly, from
the lower creek deposit.
On the other hand, according to the published accounts of
Dr. Sellards, the bones of the extinct vertebrates found in the
upper creek deposit are much scattered, commonly a few teeth, or
a lower jaw, or fragments of one or two other bones.’ In this con-
dition they do not seem to the writer to preclude more or less
reworking by the creek, but rather to imply it.
Next let us consider the location of the human bones and
artifacts. On the north bank of the canal the human relics thus
far found have come exclusively from the upper creek deposit. No
evidence of the presence of man has yet been discovered in the
lower creek deposit on the north bank of the canal. At the same
time it strikes the writer as an observation equally to be emphasized
that the two creek deposits are quite distinct from one another
throughout this section along the north bank of the canal. In this
section the observer feels little hesitation in drawing the dividing
line, and different investigators readily place it at the same level.
It can scarcely be without significance that the human relics found
thus far in the north bank of the canal all lie above this well-marked
dividing line, while the vertebrates of greatest age, and those which
present the best basis for the claim that they cannot have been
rewashed, lie below this line.
All the human relics reported to have come from the lower
creek filling were found in the south bank of the canal, and were
tE. H. Sellards, ‘‘Human Remains and Associated Fossils from the Pleistocene
of Florida,” Eighth Ann. Rept. Florida Geol. Surv., 1916, pp. 147-52.
FURTHER STUDIES AT VERO, FLORIDA 677
obtained from points marked M and N on the contour map (Fig. 4).
It was at point M that the original discovery of human bones was
made by Mr. Frank Ayers in October, 1915. But since that time
the bank at this point has been cut back five or six feet in further
search for bones, so that the exact resting-place of this first find
of bones can no longer be viewed. According to the description
given by Dr. Sellards, these human remains, or skeleton No. 1 as it
Fic. 4.—Detailed map of the locality where the human bones were found. The
canal and the resulting dump piles have done much to change the original topography.
The dotted area represents the flood-plain of Van Valkenburg’s Creek as it appears
to have been just prior to the digging of the canal. The first human skeleton was
found at point marked M, the second at point marked NV, while human relics were
found also at point O. (Reproduced from Symposium, p. 26.)
_ may be called, were imbedded in brown sand about two feet from
the surface of the ground as it existed previous to the construction
of the canal.t_ Of these two feet, nine inches, next above the bones,
consisted of brown sand, above which lay one foot three inches of
sandy, fresh-water marl. All of this was originally assigned to the
lower creek fill. If this be correct the upper filling is wanting at
t OD. cit., pp. 131-32.
678 ROLLIN T. CHAMBERLIN
this place. But in his symposium paper Dr. Sellards had come to
regard this marl as probably equivalent to formation No. 3 (upper
creek fill),t and in that article assigned to it a thickness of 18 inches.”
Only a few inches of brown sand therefore remain as a basis for
referring the bones to the lower creek deposit in a case in which
the correct reference is a matter of critical importance. The case
accordingly does not seem to the present writer to be one that may
safely be regarded as conclusive.
The other human remains reported from the lower creek deposit
were obtained in the extensive diggings carried on at the point
marked N in Fig. 4. In the section at this point the lower fill shows
extreme irregularity. This is assigned to subsequent scour and fill,
evidences of which are more marked here than anywhere else in the
sections exposed by the canal excavations. Cutting by the stream
has been so pronounced that, in the midst of the area over which
the bones are scattered, the lower deposit has at one point been
completely removed, and the upper filling rests in a depression cut
into formation No. 1 (the underlying marine beds).
A few feet to the west of this more human bones were found
along the contact line of formations Nos. 2 and 3 (the upper and
lower creek deposits), or slightly within the basal portion of the
upper creek deposit. Because of the close association of these two
finds, because there is no duplication of parts, and because all the
bones came from a large individual, Dr. Sellards believes that the
bones mentioned in the last paragraph and referred to the lower
fill and those here mentioned as found a few feet to the west along
the contact of the two fillings, all belong to the same skeleton.4
This may be called skeleton No. 2.
If these bones all belong to one skeleton, the fact that a part of
them were found in formation No. 2, as interpreted by Dr. Sellards,
and a part of them in the base of formation No. 3 requires explana-
tion. This naturally led to the suggestion that those bones which
were found in the basal portion of the upper fill reached that posi-
tion by being washed out of the lower deposit.’ If, however, one
examines Fig. 6 of the Florida state report, it is seen that the bones
t Symposium, p. 17. 3 Fighth Ann. Rept., etc., Fig. 6, p. 137.
201d pei22e 4 Tbid., p. 142. 5 Symposium, p. 54.
FURTHER STUDIES AT VERO, FLORIDA 679
found in the basal portion of the upper deposit are upstream from
the point where the bones in the lower deposit occur. Besides this,
two out of the three bones are figured as occurring at a consider-
ably higher level than the bones in the lower deposit. At the same
time the attitude and appearance of these suggest that they had
already moved somewhat down the rather steep slope implied by
the depositional lines.
These suggestive relations occur at the most critical locality.
It was here that most of the collecting was done, not only during
this later visit, but also during the previous one. From the geo-
logical point of view this section is peculiar in that here there has
been more obvious scour and fill by the stream than elsewhere.
This is made evident by an unusual number of pockets and lenses
of sand and muck, as well as rapid dovetailings of layers. It may
be worthy of note also that the section here lies beneath the latest
channel of Van Valkenburg’s Creek. The pockets, ‘filled holes,”
lenses, and dovetailings render the identification of the true line
between the lower and upper creek deposits both difficult and
uncertain. While the line of division is reasonably distinct at most
points elsewhere, as on the north bank already noted, it unfortu-
nately becomes obscure in this critical section.
In the course of our examinations there frequently arose ques-
tions as to the line of division between the upper and lower deposits,
and sooner or later the judgments of all members of the party were
more or less involved in these efforts at discrimination. These
questions revealed the fact that there were notable differences of
opinion as to whether a given bit of a section belonged to the upper
or lower deposit. If, as discussion and critical consideration pro-
ceeded, there was noticeable a tendency to shift the dividing line
in one direction rather than another, it was to give the base of the
upper creek filling a lower place in holes and hollows than it had
been assigned before. In other words, there was a general dispo-
sition, as the result of progressive study, to lower the division line.
This justifies the inference that any sharp division of the creek
deposits in this portion of the south bank into distinct formations
is lacking in complete conclusiveness.
680 ROLLIN T. CHAMBERLIN
A specific case of this kind of uncertainty is illustrated by Fig. 5.
As a result of much digging for fossils at this point, the bank had
gradually been cut back till, at the taking of this picture, it was
perhaps 15 feet back from its original canal face. The entire thick-
ness of the deposit from the base of the black muck and sand fill
Fic. 5.—Section of south bank of the canal near point marked NW (Fig. 4) as
exposed by the party on March 21, 1917. At the base are buff marine sands with
some shells (formation No. 1); No..2 above is the lower creek fill which was originally
supposed to extend up to the prominent line of white sand lenses, just beneath
marker 3 in the middle of the picture; but upon more critical inspection, the patches
of fill marked 3a and 30 were excluded from the lower fill and placed in the upper
fill. No. 3 represents the unmistakable upper creek fill. It contains some small
lenses and pockets of marine shells derived from formation No. 1.
(just to the left of the hammer) to the surface as it was in 1913, just
prior to the excavation of the canal, is 5} feet. As the party viewed
the newly exposed section for the first time, all were ready to carry
the upper creek filling down to the prominent line of whitish sands
and reworked coquina shells just beneath marker 3 in Fig. 5. After
a brief inspection, there seemed to be reasons for assigning the block
FURTHER STUDIES AT VERO, FLORIDA 681
marked 3a (Fig. 5) to the upper creek deposit. And then the upper
deposit was extended downward to include the peculiar funnel-like
fill marked 3), while the writer, at least, would hesitate to deny
that the upper deposit might not, in reality, include also some of
the material which reposes in lenslike fashion adjacent to this
funnel. In any case it is clear that there was much scouring and
filling at this point, and this involved the lower as well as the
upper deposit. This suggests that the scour and fill arose from the
course of the stream at this point—some turn, perhaps, or some
configuration of its channel.
The peculiar funnel-like filling marked 36 was so obvious as
to suggest the.name “funnel,” as it was evidently a deep hole in
the creek bed filled with alluvium. After the photographs were
taken, further excavations were made, and at the bottom of the
funnel the carapaces of two turtles were found. One of these,
still firm and strong, has already been referred to (p. 676).
With further horizontal digging into the bank the funnel quickly
disappeared.
This particular locality has been a gold mine for bone-collecting,
and far more excavating has been done here than at any other point.
The writer suspects that one reason why this particular area has
proved so prolific in results is that there was an exceptional rework-
ing of material by the-stream at this point, resulting in a greater
concentration of bones, pottery, and coarser material. At the same
time, this material was left in more unusual positions than in places
where the stream action has been simpler. In fact, small lenses
and stringers of shells derived from the erosion of the underlying
marine coquina are frequently seen here, not only in the lower
creek deposit, but in the upper creek deposit as well. If, then, the
upper creek deposit has received an appreciable portion of its
material from the more deeply buried marine beds, how much more
of its material must have been derived from the far more accessible
lower creek deposits which overlie these marine beds. The mixing
of materials is obvious.
In view of the similarity of the upper and lower creek deposits,
and the inevitable difficulty of drawing a perfect line of division
between them; in view of the actual differences of opinion as to
682 ROLLIN T. CHAMBERLIN
just where such line should be drawn, and of changes of opinion
once formed; in view of the natural doubt as to whether two such
deposits measuring together only about six feet could in fact remain
altogether unmixed and distinct; and in view of the observed fact
that the stream, in its later action, actually did cut entirely through
its own earlier deposits and into the marine formation below, it
would seem that grave doubts as to the trustworthiness of correla-
tions of this stream material may well be entertained. Perhaps it
is obligatory that they should be entertained. The balance of evi-
dence seems to lie in favor of including all doubtful horizons in the
upper fill, since the upper fill does penetrate deep into the lower
fill at so many points. The human bones and relics would seem
to the present writer to belong to the upper creek deposit, which
was contemporaneous with the. human occupation of Florida.
This interpretation would allow the correctness of Dr. Sellards’
contention that the bones of the extinct vertebrates well down in
the undisturbed part of the lower creek deposit are fossils primary
to that deposit. With this revision of the stratigraphic view, the
testimony of the inherent character of the human relics rises into
scarcely less than decisive importance.
Now, among the human relics, the pottery seems to carry the
most telling testimony as to the time when the aborigines dwelt
on the banks of Van Valkenburg’s Creek. The association of the
pottery with the human bones may well be regarded as peculiarly
significant, for the pottery was a human product and it carries a
time relation. Fragments of pottery, in more or less abundance,
were found on the second visit at as low a horizon in the creek
deposits as were any of the human bones. The writer saw no evi-
dence of any human race earlier than the pottery-makers, and no
such earlier race has been claimed. Now, as MacCurdy,* Hrdlicka,?
and Holmes? have pointed out, the pottery belongs to the type which
was used by the mound-building Indian tribes of Florida. Such
pottery was in common use in the middle or later Neolithic age.
This pottery, of itself, would not therefore be assigned a date
earlier than mid-Recent. Even in Europe, where the presence and
t Symposium, pp. 60-62.
2 [bid., pp. 47-50. 3 [bid., p. 51.
FURTHER STUDIES AT VERO, FLORIDA 683
development of man has been traced from the mid-Pleistocene on,
the introduction of pottery by Neolithic man is not placed as far
back as the close of the Glacial period and is not, therefore, Pleisto-
cene as usually defined. There is no ground to suppose that
pottery was in use in North America before it was in use in the
Old World; more probably it was introduced here later.
If (1) the testimony of the human relics, particularly that of
the pottery, be taken at its apparent paleontological value; if
(2) the upper creek fill, whose accumulations demonstrably con-
tinued on until 1913, be regarded as embracing all the human
relics, as seems quite consistent with the physical evidence; if
(3) the critical extinct vertebrate fossils found in this upper creek
fill be regarded as derivatives from the lower creek fill; and if
(4) the lower creek fill be regarded as contemporaneous with the
last living stages of the extinct vertebrates whose fossils it holds
as primary inclusions, as Dr. Sellards contends, the whole history
becomes consistent physically and paleontologically, and the gist
of its lesson is that the Pleistocene fauna lived longer in this
genial southern clime than it has been credited with in the more
northern latitudes, while the evidence of man’s presence here falls
into harmony with the general tenor of other evidences which fail
to assign him an antiquity beyond the mid-Recent.
ANOTHER LOCALITY OF EOCENE GLACIATION IN
| SOUTHERN COLORADO’
WALLACE W. ATWOOD
Harvard University, Cambridge, Massachusetts
Since the publication of the paper on the Eocene glaciation
recorded at the northwest base of the San Juan Mountains near the
village of Ridgway,’ the author’s attention has been called to a
similar discovery made by Mr. Charles W. Drysdale in British
Columbia at about the same time.’
When Eocene till was found near Ridgway, and the formation
was given the name Ridgway till, it was anticipated that other
glacial deposits of the same age would soon be recognized in other
parts of the Rocky Mountain province. Each of the larger ranges
in this great geographic province has had a history somewhat
similar to that of the San Juan Mountains. These ranges were
all uplifted, some as great anticlinal arches, some as domes, and
others with some faulting and intrusion, at the close of the Meso-
zoic era or beginning of the Cenozoic time. Those great arches
and domes were dissected into mountain forms, and, when favorable
climatic conditions prevailed, glaciers probably formed in many
of the higher basins among those mountains and assisted in the
further dissection of the ranges. Now that Eocene till has been
discovered in British Columbia, and at a locality to be herein
described near the south margin of the San Juan Mountains, it
appears to be well established that conditions favorable for the
formation of Alpine glaciers did obtain in the western portion of
North America during early Tertiary time.
1 Published with the permission of the Director of the United States Geological
Survey.
2W. W. Atwood, ‘“‘Eocene Glacial Deposits in Southwestern Colorado,” U.S.
Geol. Survey, Prof. Paper 95-B, 1915.
3C. W. Drysdale, “Geology of Franklin Mining Camp, British Columbia,”
Canadian Geol. Survey Mem. 56, 1915.
684
EOCENE GLACIATION IN SOUTHERN COLORADO 685
The locality at which this most recent discovery of Eocene till
in Colorado was made is about 20 miles southeast from Pagosa
Springs and in the south-central portion of the Summitville quad-
rangle of the United States topographic atlas.
The deposit is exposed in the valley walls of White Creek where
that stream is dissecting the surface of V Mountain. The best
exposures may be reached by trail from the Blanco Basin, following
the base of the bold mountain escarpment just east of V Mountain
to a large lake held in by recent landslides, and thence westward
half a mile to the junction of the two upper forks of White Creek.
The ridge between the two upper forks of White Creek and that
west of the west fork are composed of this ancient till, but on their
surfaces there are fragments of the later Tertiary volcanics that
have fallen or been washed from the mountains to the east.
The till is composed of stones ranging up to 5 feet in diameter
imbedded in a clay matrix. Many of the stones are distinctly
striated, and most of them are subangular and beautifully polished
and planated. The notable character of this till, however, is the
abundance of stones that have come from the pre-Cambrian forma-
tions, now nowhere exposed near this locality, and the many bowl-
ders known to have come from the Cutler or Dolores formations
of Permo-Triassic age which must also be buried in the core of the
range. Of equal significance is the absence of stones from the later
Tertiary volcanics. These two points make it clear that the ice
which deposited this till formed and accomplished its work during
the time when the pre-Cambrian core of the range and the upturned
Paleozoic and Mesozoic formations were exposed at the surface,
and before the later Tertiary lavas and tuffs were present.
The stones in this till consist of granites, quartz, quartzites,
schists, gneisses, jaspers, red sandstones from the Cutler or Dolores
formations, and conglomerates from one or the other of those
formations. There are also many porphyries and some bowlders
of a tuff-breccia, just as there are in the type section of the Ridgway
till. These igneous and volcanic rocks were derived from an earlier
series of intrusives and eruptives and are quite distinct in age from
the later volcanics which constitute the mass of the present moun-
tains.
686 WALLACE W. ATWOOD
The lithological character of this drift is distinctly different
from that of the three Pleistocene drift deposits which are so com-
monly found in the foothill regions bordering the San Juan Moun-
tains, and which are all present in this immediate district. The
Pleistocene glacial deposits are characterized by the stones of the
later Tertiary volcanics and usually contain very little that could
not have been derived from those volcanics.
In one exposure near the junction of the two forks of White
Creek on V Mountain a pebble-clay till is exposed which resembles
the upper member of the Ridgway till at the type locality. This
pebble clay contains many stones less than one-quarter of an inch
in diameter and a few cobblestones and small bowlders. The best
striae were found on stones that were taken from this pebble-clay
phase of the till.
Beneath this exposure of Eocene till is the Mancos shale, and
in this respect the conditions are identical with those at the north-
west'base of the range. Upstream from the best exposures of the
till an andesitic rock cuts the Mancos shale and appears to be at
the base of the till for some little distance. On the slopes above
this deposit of till there are beautifully waterworn pebbles of pre-
Cambrian rocks similar to those that characterize the Eocene
glacial deposit. They appear to have come from the complete
disintegration of a conglomerate. Such a conglomerate overlies
the Ridgway till at almost all of the known localities.
This section is somewhat less satisfactory than many of those
described in the first report on Eocene glaciation in the San Juan
Mountains, for it is not at present overlain by the later Tertiary
volcanics. The lithologic character of this deposit determines
its age.
REVIEWS
Eocene Glacial Deposits of Southwestern Colorado. By WALLACE
W. Atwoop. Prof. Paper, U.S. Geol. Surv., No. 95-B, 1915.
Pp. 13-26, pls. 4, figs. 11.
Glacial deposits of Eocene age were discovered in 1913 near Ridg-
way, Colorado, northwest of the San Juan Mountains. The nine
exposures are scattered over an area of 20 square miles. The Ridgway
till rests on the Mancos shale and is overlain by the Telluride con-
glomerate and San Juan tuff. The till is divided into two members.
The lower is a bowlder till containing many striated stones, some very
large. The upper till is a dark slate-colored clay, unstratified and
containing only a few striated pebbles. The bowlder till is believed to
have been deposited by glaciers heading in the region of the present
San Juans. The pebble till may have been deposited by ice moving
over extensive surface exposures of Mancos shale from the region of
the West Elk Mountains to the northeast.
The paper closes with a summary of the distribution of pre-
Pleistocene glaciation. An extensive bibliography is appended.
Ho RB.
The Yenina District, Alaska. By STEPHEN R. Capps. USS.
Geol. Surv., Bull. No. 534, 1913. Pp. 75, pls. 13, figs. 7.
This area lies along the southeast base of the Alaska Range in the
drainage basin of the Yentna River, a tributary of the Susitna. The
oldest rocks are a pre-Jurassic series of slates and graywackes. They
are everywhere faulted and folded, and are intruded by igneous rocks
ranging from granite to diorite. The intrusives are provisionally
assigned to the late Lower Jurassic or Middle Jurassic. Older dikes
of diabase and greenstone have been deformed and metamorphosed
along with the slate series.
Next younger are rocks of Eocene age, consisting of sands, shales,
gravels, and commonly some lignitic coal. Coarse stream gravels
overlie the coal-bearing series. Evidence of the Tertiary age of the
gravels was obtained. They were formerly regarded as Pleistocene.
687
688 REVIEWS
They are probably equivalent to the Nenana gravels on the north side
of the Alaska Range. i
Pleistocene and recent glaciation are described. The present snow
fields and glaciers are confined to the heads of the valleys and the higher
portions of the Alaska Range.
Placer gold was first discovered in the Yentna district in 1905.
The total production up to 1911 was $383,000. ‘The gold is believed
to have been derived from quartz veins and stringers associated with
the intrusions in the slate and graywacke series.
The coal, in beds ranging from 3 to 12 feet in thickness, is a medium-
to low-grade lignite.
H.R. B.
RECENT PUBLICATIONS
—Jahresberichte und Mitteilungen des Oberrheinischen geologischen Vereines.
Neue Folge. Bd. 4. Jahrgang 1914. Heft 2. Unter der Schriftleitung
der jeweiligen Schriftfiihrer, zur Zeit des Professor Dr. Wilhelm Solomon
und des Rechnungsrates Dr. D. Hiaberle in Heidelberg. Juli, ror.
In Kommission bei der E. Schweizerbart’schen Verlagsbuchhandlung
Nagele und Dr. Sproesser. [Stuttgart, 1914.]
—Jahresberichte und Mitteilungen des Oberrheinischen geologischen Vereines.
Neue Folge. Bd. 5. Jahrgang 1915. Heft 1. Unter der Schriftleitung
der jeweiligen Schriftfiihrer, zur Zeit des Professor Dr. Wilhelm Solomon
und des Rechnungsrates Dr. D. Hiaberle in Heidelberg. November, 1915.
—Jounson, R. H., anp Hunttey, L. G. Principles of Oil and Gas Produc-
tion. [New York: John Wiley & Sons, 1016.]
—Katz, F.J. Abrasive Materials in 1915. [U.S. Geological Survey, Mineral
Resources of the United States, 1915, Part II, pp. 65-80. Washington,
1916.]
Feldspar in 1915. [U.S. Geological Survey, Mineral Resources of
the United States, 1915, Part II, pp. 43-53. Washington, 1016.]
Silica in 1915. [U.S. Geological Survey, Mineral Resources of the
United States, 1915, Part II, pp. 55-60. Washington, 1916.]
—Kay, F. H., anp WuitE, K. D. Coal Resources of District VIII (Dan-
ville). Illinois Coal Mining Investigations, Co-operative Agreement,
Bulletin 14. State Geological Survey, Engineering Experiment Station,
University of Illinois; U.S. Bureau of Mines. [Illinois Geological Sur-
vey. Urbana, 1015.]
—Kuinpie, E. M. Notes on the Geology and Palaeontology of the Lower
Saskatchewan River Valley. [Canada Department of Mines, Museum
Bulletin 21, No. 1576, Geological Survey, Geological Series, No. 30.
Ottawa, October 14, 1915.|
—Kress, C. E., aided by TEEts, D. D., JR. West Virginia County Reports,
1916. Raleigh County and Western Portions of Mercer and Summers
Counties. With maps showing topography of Mercer County, topography
of Raleigh County, and western part of Summers County, general and
economic geology of Raleigh County and western part of Summers County,
and general and economic geology of coal area only in Mercer County.
[West Virginia Geological Survey. Morgantown, 1016.]
—LangE, F. K. Annual Report of the Secretary of the Interior for the Fiscal
Year Ended June 30, 1915. [Washington, 1915.]
6890
690 RECENT PUBLICATIONS
—LAtimer, W. J. Soil Survey of Raleigh County, West Virginia. Advance
Sheets—Field Operations of the Bureau of Soils, 1914. [U.S. Depart-
ment of Agriculture, Bureau of Soils. Washington, 1916.]
—LeitH, C. K., AnD Meap, W.J. Metamorphic Geology, A Textbook. [New
York: Henry Holt & Co., 1915.]
—LIiESEGANG, R.E. Die Achate. [Dresden und Leipzig: Verlag von Theodor
Steinkopff, 1915.]
—Lisson, C. I. Edad de los Fosiles Peruanos. Dos Indices de Consulta.
[Lima: Imprenta de la Escuela de Ingenieros, J. Mesinas, 1915.]
—Matcorm, W. The Oil and Gas Fields of Ontario and Quebec. [Canada
Department of Mines, Memoir 81, No. 1561, Geological Survey, Geological
Series, No. 67. Ottawa, 1915.]
—MarsHAtLt, R. B. Profile Surveys in Chelan and Methow River Basins,
Washington. [U.S. Geological Survey, Water-Supply Paper 376. (Pre-
pared in co-operation with the State of Washington.) Washington, 1015.]
Profile Surveys in Spokane River Basin, Washington, and John
Day River Basin, Oregon. [U.S. Geological Survey, Water-Supply
Paper 377. Washington, 1915.|
Profile Surveys in 1914 in Umpqua River Basin, Oregon. [U.S.
Geological Survey, Water-Supply Paper 379. (Prepared in co-operation
with the State of Oregon.) Washington, 1915.]
Profile Surveys in 1914 on Middle Fork of Willamette River and
White River, Oregon. [U.S. Geological Survey, Water-Supply Paper 378.
(Prepared in co-operation with the State of Oregon.) Washington, ror5.|
Spirit Leveling in West Virginia, 1896 to 1015, Inclusive. [U.S.
Geological Survey, Bulletin 632. Washington, 1916.]
—Maryland Geological Survey. Upper Cretaceous (2 vols.) [Baltimore:
Johns Hopkins Press, 1916.]
—Michigan Geological and Biological Survey, Publication 18, Geological
Series 15. Contributions to the Pre-Cambrian Geology of Northern
Michigan and Wisconsin, by R. C. ALLEN and L. P. BARRETT. The
Geology of Limestone Mountain and Sherman Hill in Houghton County,
Michigan, by E. C. Case and W. I. Roprnson. [Lansing, 1915.]
—Michigan Geological and Biological Survey, Publication 19, Geological
Series 16. Mineral Resources of Michigan, with Statistical Tables of Pro-
duction and Value of Mineral Products for 1914 and Prior Years. Witha
Treatise on Michigan Copper Deposits, by R. E. Horr. ([Lansing, 1915.]
—NMineralchemie, Handbuch der. Bd. II. 9 (Bog. 31-40). [Dresden und
Leipzig: Verlag von Theodor Steinkopff, 1915.]
—Mines and Metallurgy, University of Missouri, School of. Bulletin,
November, 1915, Technical Series. [Rolla, 1915.]
—Mining Operations in the Province of Quebec during the Year 1914, Report
on. [Mines Branch, Department of Colonization, Mines, and Fisheries of
the Province of Quebec, Canada. Quebec, 1915.]
RECENT PUBLICATIONS 691
—Missouri School of Mines and Metallurgy, University of Missouri, Bulletin,
Vol. II, No. 3. Cupellation Losses in Assaying. [Rolla, February, 1916.]
—Morean, P. G. Eighth Annual Report (New Series) of the Geological
Survey Branch of the Mines Department, Appendix C. [New Zealand
Geological Survey. Wellington, rors.]
—Morean, P.G., anD BARTRUM, J. A. The Geology and Mineral Resources of
the Buller-Mokihinui Subdivision, Western Division. Bulletin 17 (New
Series), Geological Survey Branch, New Zealand Department of Mines:
[Wellington, ror15.]
—Murpvocu, J. Microscopical Determination of the Opaque Minerals. An
Aid to the Study of Ores. [New York: John Wiley & Sons, 1916.]
—Museu Nacional do Rio de Janeiro, Archivos do. Vol. XIX. [Rio de
Janeiro: Imprensa Nacional, 1916.]
—New Zealand Geological Survey Branch, Ninth Annual Report (New Series)
of the. Appendix C. [Wellington, ro15.]
—Osporn, H. F. Close of Jurassic and Opening of Cretaceous Time in North
America. Proceedings of the Paleontological Society. Bulletin of the
Geological Society of America, Vol. XXVI, 295-302. August 17, 1915.
Review of the Pleistocene of Europe, Asia, and Northern Africa.
Annals of New York Academy of Sciences, Vol. XXVI, 215-315. [New
York, 1015.]
—Pahasapa Quarterly, The. February, 1916. Tungsten Number. [South
Dakota School of Mines, Rapid City, 1916.]
—PocvE, J. E. The Emerald Deposits of Muzo, Colombia. [Preprint from
Transactions of American Institute of Mining Engineers, 1916.]
—Recer, D. B. Detailed Report on Lewis and Gilmer Counties. Accom-
panied by topographic and geologic maps. [West Virginia Geological
Survey. Morgantown, 1916.]
—REINECKE, L. Ore Deposits of the Beaverdell Map-Area. [Canada
Department of Mines, Memoir 79, No. 1537, Geological Survey, Geologi-
cal Series, No. 65. Ottawa, 1o15.]
—Ricuarps, H. C. The Volcanic Rocks of South-Eastern Queensland.
Proc. Roy. Soc. of Queensland, Vol. XXVII, No. 7. [Brisbane, March,
1916.]
—kRtes, Hetnricu. Economic Geology. 4th ed. [New York: John Wiley
& Sons, 1916.]
—Ropsinson, H. M. Ozokerite in Central Utah. [U.S. Geological Survey,
Bulletin 641-A. Washington, 1916.]
—RussELL, E. J. A Student’s Book on Soils and Manures. Cambridge
Farm Institute Series. [London: Cambridge University Press, Fetter
Lane, E. C. New York: Putnam.]
—Saunp_Ers, E. J. The Coal Fields of Kittitas County. [Washington Geo-
logical Survey, Bulleting. Olympia, 1914.]
692 RECENT PUBLICATIONS
—School of Mines and Metallurgy, University of Missouri, Bulletin, Vol. VIII,
No. 1. Bibliography, Concentrating Ores by Flotation. [Rolla, Janu-
ary, 1916.]
Vol. VIII, No. 2. Catalogue, 1915-1916. [Rolla, March, 1916.]
—Section Géologique du Cabinet de sa Majesté, Travaux de la. (Ministére
de la Maison de ’Empereur.) Vol. VIII, Livr. II. [Petrograd, 1915.]
—Seismological Society of America, Bulletin of. Vol. 5, No. 3. [Stanford
University, California, 1915.]
Vol. 5, No. 4. [Stanford University, California, r915.]
Vol. 6, No. 1. [Stanford University, California, 1916.]
—SELLARDS, E. H. The Pebble Phosphates of Florida. Seventh Annual
Report, Florida Geological Survey, pp. 25-116. [Tallahassee, ro15.]
—SHIMER, H. W. An Introduction to the Study of Fossils. (Plants and
Animals.) [New York: Macmillan, 1914.]
—South Dakota School of Mines. The Pahasapa Quarterly, Vol: V, No. 1.
[Rapid City, December, 1915.]
—STAUFFER, C. R. The Devonian of Southwestern Ontario. [Canada
Department of Mines, Memoir 34, No. 1248, Geological Survey, Geological
Series, No. 63. Ottawa, 1915.]-
—STEPHENSON, L. W., AND CripER, A. F. Geology and Ground Waters of
Northeastern Arkansas. With a discussion of the Chemical Character
of the Waters by RicHarp B. Dore. [U.S. Geological Survey, Depart-
ment of the Interior, Water-Supply Paper 399. Washington, 1916.]
—STEVENS, G. C. Surface Water Supply of Virginia. [Virginia Geological
Survey, Bulletin X. Charlottesville, 1916.]
—SutTer, H. Alphabetical Hand-List of New Zealand Tertiary Mollusca.
[New Zealand Geological Survey. Wellington, 1915.]
Revision of the Tertiary Mollusca of New Zealand, Based on Type
Material. Part II. New Zealand Geological Survey, Palaeontological
Bulletin No. 3. [New Zealand Department of Mines. Wellington, 1o15.]
—Warp,L. K. Annual Report of the Government Geologist of South Aus-
tralia, for 1914, Together with an Appendix Comprising a Catalogue of
Official Publications Dealing with the Geology and Mineral Resources
of South Australia. With Plan. [Adelaide, 1o015.]
—Washington Academy of Science, Journal of. Vol. V, Nos. 14, 16, 17, 18,
19, 20, 21. [Baltimore: Waverly Press, 1915.] Vol. VI, Nos. 1-13.
[Baltimore: Waverly Press, 1916.]
—Wartson, T. L. Administrative Report of the State Geologist for the
Biennial Period 1914-1915. [Virginia Geological Survey. Charlottes-
ville, 1916.]
—WELILER, S. (1) Atractocrinus, a New Crinoid Genus from the Richmond
of Illinois. (2) Description of a Ste. Genevieve Limestone Fauna from
Monroe County, Illinois. Contributions from Walker Museum, Vol. I,
No. 10. [Chicago: The University of Chicago Press, 1916.]
WRITERS, EDITORS, BUSINESS MEN
Stenographers, Teachers, Librarians, Ministers, Lawyers, Publishers, Proofreaders
SHOULD USE THE TWO BOOKS DESCRIBED BELOW AS
CONSTANTLY AS THEY DO THEIR HAND DICTIONARIES
The fifth revised edition of
A MANUAL OF STYLE
(READY NOVEMBER 10)
By the Staff of the University of Chicago Press
is for all whose work involves.typography and contains, among other things,
One hundred eighteen pages of Rules for Composition.
An entire chapter devoted to Technical Terms.
A page illustrating Proofreader’s Marks.
An appendix of valuable Hints to authors and editors, proofreaders, and copyholders.
A full and comprehensive Index.
One hundred fifty pages illustrating Specimens of Types in Use at the University of
Chicago Press.
“The best book of the kind published.”—The Independent
Cloth bound—$r1.50, postage extra
A companion volume to A Manual of Style is
A MANUAL FOR WRITERS
By Joun M. Manty
Head of the Department of English in the University of Chicago
and Joes A. POWELL
This book is designed to aid all who are concerned with the writing of English. It
_ aims to answer the practical questions that constantly arise in the everyday experience
of individual writers, business houses, schools and colleges, editors, secretaries, etc. The
table of contents suggests the practical usefulness of the book.
“I use it as I do my hand dictionary.”—BERTRAM L. JONES, Head of the Depart-
ment of English, Western State Normal School, Kalamazoo, Michigan.
Cloth bound—$1.25, postage exira
THE UNIVERSITY OF CHICAGO PRESS
CHICAGO 5825 Ellis Avenue ILLINOIS
The Maximum in Microscopes -
Whether for class or laboratory work, only
the highest optical accuracy and mechani-
cal efficiency should be sought when
purchasing microscopes. If instruments
are being sought that are ideally suited to
give long strenuous usage as well as
dependable results, the varied line of
Rausch lomb
Microscopes
should be given careful consideration. These in-
struments predominate inthe country’s laboratories,
because their quality has become thoroughly estab-
lished through many years of scientific and pro-
duciive effort along optical lines.
Our microscopes, specially adapted for school
use, are reasonably priced, and their precision un-
questioned.
Consider, for instance, Model F-2 here illustrated.
This instrument meets all class requirements of
school and college. Long, curved handle arm to
provide space for manipulation of object and built
to “‘stand up” under constant usage. This model.
is priced at $31.50. Others range from $18.00
up. Write for a copy of our catalog or descriptive
circular.
Special Terms to Educational Institutions
Bausch & lomb Optical CG.
654 ST. PAUL STREET ROCHESTER, N.Y
New York Washington Chicago
Leading American Makers of Microscopes, Projection Apparatus,
Photographic Lenses, Ophthalmic Lenses and Instruments, Stereo-
Prism Binoculars, and other High-Grade Optical Products.
San Francisco
SELF STARTING
REMINGTON
TYPEWRITER
| Saves from 15% to 25% of
every letter-writing hour
IS new invention permits your
typist to keep her eyes onher copy.
The machinedoesn’t have to be looked
at, or the scale watched. The time
saving isautomatic. Thereis no other
typewriter like this. Fully protected
by Remington patents.
Descriptive folders mailed on request.
REMINGTON TYPEWRITER CO.
(Incorporated)
NEW YORK AND EVERYWHERE »
FINE INKS 4x» ADHESIVES
For those who KNOW
Drawing Inks
Eternal Writing Ink
a Engrossing Ink -
e ) 9 Taurine Mucilage ;
Hig Gims? ( proto mounter'Paste
-) Drawing Board Paste
Liquid Paste
Office Paste
Vegetable Glue, Etc.
Are the Finest and Best Inks and Adhesives
Emancipate yourself from the use of corrosive and
ill-smelling inks and adhesives and adopt the
Higgins Inks and Adhesives... They will bea
revelation to you, they are so sweet, clean, well
put up, and withal so efficient.
At Dealers Generally
CHAS. M. HIGGINS & CO., Mfrs.
Branches: Chicago, London
271 Ninth Street Brooklyn, N.Y.
=
» VOLUME XXV i NUMBER 8
THE
OURNAL or GEOLOGY
A SEMI-QUARTERLY
EDITED BY
' THOMAS C. CHAMBERLIN AND ROLLIN D. SALISBURY
With the Active Collaboration of
SAMUEL W. WILLISTON, Vertebrate Paleontology ALBERT JOHANNSEN, Petrology
STUART WELLER, Invertebrate Paleontology ROLLIN T. CHAMBERLIN, Dynamic Geology
ALBERT D. BROKAW, Economic Geology
ASSOCIATE EDITORS
SIR ARCHIBALD GEIKIE, Great Britain JOSEPH P. IDDINGS, Washington, D.C.
CHARLES BARROIS, France JOHN C. BRANNER, Leland Stanford Junior University
ALBRECHT PENCK, Germany RICHARD A. F. PENROSE, Jr., Philadelphia, Pa.
Pe Noe WILLIAM H. HOBBS, University of Michigan
__T. W. EDGEWORTH DAVID, Australia ‘ PROEINUS Ds ASIDE WUtSy MEG CHL Uh CUS 5)
_ BAILEY WILLIS, Leland Stanford Junior University (ENE ESS SEND) UK a MOD Eset
GROVE K. GILBERT, Washington, D.C. WALLACE W. ATWOOD, Harvard University
CHARLES D. WALCOTT, Smithsonian Institution t WILLIAM H. EMMONS, University of Minnesota
_ HENRY S. WILLIAMS, Cornell University . ARTHUR L. DAY, Carnegie Institution
NOVEMBER-DECEMBER 1917
_ THE ACTIVE VOLCANOES OF NEW ZEALAND- - - - - - E.S. Moore 693
FOOTHILLS STRUCTURE IN NORTHERN COLORADO~ - - - Victor ZIEGLER 715
ON THE GEOLOGY OF THE ALKALI ROCKS IN THE TRANSVAAL H. A. BROUWER 741
PETROLOGICAL ABSTRACTS AND REVIEWS” - otis - .- ALBERT JOHANNSEN 7709
RUINED SM eT a pci wictiucx nk eandpiictowrey >) SBronn Ber helt EN) NIC i i
RECENT PUBLICATIONS ssa beers tie eet aR a es a a ea i Ne CR)
: {
THSTD ESS. {TRO NOTE GING OS Vea RE 5 ac RR Me eng an a oR meee me EO Peet)
PB WINNERS th OF CHICAGO PRED?
CHICAGO, ILLINOIS, U.S.A.
———
HS OniAM Insis ih ©
“2 \
Jie Iddings
3 |
COLLECTION
ai. ve
“Ationn| mused™
: THE CAMBRIDGE UNIVERSITY PRESS, Lonpon anp EpINBURGH
3 THE MARUZEN-KABUSHIKI-KAISHA, Toxyo, Osaka, Kyoto, FukKuOKA, SENDAI
THE MISSION BOOK COMPANY, SHANGHAI
THE JOURNAE- OF GEOLOGY
EDITED BY
THOMAS C. CHAMBERLIN AND ROLLIN D. SALISBURY
With the Active Collaboration of
SAMUEL W. WILLISTON : ALBERT JOHANNSEN
Vertebrate Paleontology Petrology
STUART WELLER ROLLIN T. CHAMBERLIN
Invertebrate Paleontology Dynamic Geology
ALBERT D. BROKAW
Economic Geology
The Journal of Geology is published by the University of Chicago at the University Press semi-
quarterly, on or about the following dates: February 1, March 15, May 1, June 15, August 1, September 15,
November 1, December 15. { The subscription price is $4.00 per year; the price of single copies is
65 cents. Orders for service of less than a half-year will be charged at the single-copy rate. | Postage is
prepaid by the publishers on all orders from the United States, Mexico, Cuba, Porto Rico, Panama Canal
Zone, Republic of Panama, Hawaiian Islands, Philippine Islands, Guam, Samoan Tslanee Shanghai.
{ Postage is charged extra as follows: For Canada, 30 cents on annual subscriptions (total $4.30), on
single copies 4 cents (total 69 cents); for all other countries in the Postal Union, 53 cents on annual sub-
scriptions (total $4.53), on single copies 11 cents (total 76 cents.) {| Patrons are requested to make all
remittances payable to The University of Chicago Press in postal or express money orders or bank drafts.
The following are authorized to quote the prices indicated:
For the British Empire: The Cambridge University Press, Fetter Lane, London, E.C. 4. Yearly
subscriptions, including postage, 18s. 8d. each; single copies, including postage, 3s. 2d. each.
For Japan and Korea: The Maruzen-Kabushiki-Kaisha, rr to 16 Nihonbashi Tori Sanchome,
Tokyo,. Japan. Yearly subscriptions, including postage, Yen 9.00 each; single copies, including
postage, Yen 1.50 each.
For China: The Mission Book Company, 18 Peking Road, Shanghai. Yearly subscriptions
$4.00; single copies 65 cents, or their equivalents in Chinese money. Postage extra, if mailed direct
outside of Shanghai, on yearly subscriptions 53 cents, on single copies 11 cents.
Claims for missing numbers should be made within the month following the regular month of publi-
cation. The publishers expect to supply missing numbers free only when losses have been sustained in
transit, and when the reserve stock will permit. Deliveries are not guaranteed within the war zones of
Europe, nor in other countries where transportation is disturbed by conditions incident to war.
Business correspondence should be addressed to The University of Chicago Press, Chicago, Ill.
Communications for the editors and manuscripts should be eu Esa to the Editors of THE
JouRNAL or GEotoey, the University of Chicago, Chicago, Ill.
Contributors are requested to write scientific and proper names with particular care and in citations
to follow the form shown in the pages of the Journal.
Separates, if desired, must be ordered in advance of publication. Separates of leading articles will be
printed as ordered, of which 50 (without covers) will be furnished gratis, the actual cost of the remainder
(and covers, if desired) to be paid for by the author. The actual cost may vary, and will depend upon
the amount of work in remaking the pages into forms, presswork, paper, binding, etc. Separates con-
taining half-tones may be expected to cost somewhat more, the increase depending upon the number
of cuts and the amount of work required upon them.
Entered as second-class matter, March 20, 1893, at the Post-office at Chicago, IIl., under the Act of March 6 1879
VOLUME XXV NUMBER 8
THE
feoORN AL OF CEOLOGY
NOVEMBER-DECEMBER 1917
THE ACTIVE VOLCANOES OF NEW ZEALAND
E. S. MOORE
State College, Pennsylvania
The northern island of New Zealand has, at the present time,
five volcanoes which show more or less activity, besides a large
number of others which have been active since Miocene time and
are now dormant or extinct. This island has experienced much
more volcanism during late geological time than the southern island,
which consists largely of sedimentary and ancient metamorphic
rocks. After traveling through North Island the writer was
impressed by the simple statement of the Maori guide living near
Mount Tarawera, who said, ‘‘ New Zealand has been turned over
and over.”
The active volcanoes are White Island, in the Bay of Plenty,
which displayed fresh activity in the autumn of 1914; Tarawera,
near Rotorua, which suffered a terrific explosion in 1886; Ruapehu,
which is in the solfataric stage and almost extinct; Ngauruhoe and
Tongariro, which are in the solfataric stage, but still suffer explosive
‘outbursts, those of Ngauruhoe being of considerable violence at
times. The three last-named volcanoes are situated close together
on the plateau in the central portion of the island.
There seems to be a close relationship among all these five
volcanoes, as they are arranged along an almost direct line, indicat-
ing a zone of fissuring of immense proportions, known as the
693
604 E. S. MOORE
Whakatane fault. Speight considers that this line continues from
Ruapehu through Tonga and Samoa toward Hawaii along what he
(72 74 176 (76
WY
ys EA lt rh yo
Or ) © Isuaae
WS 7é Ahvuroa
Tang gariKe
(6956)
ate pop )
Nw zE he
Gar. Egan
aS. ik 6260’ ie «Rug,
iN <
PM a8)
MAP. OF
NORTH ISLAND
NEW ZEALAND
SHOWING THE ZONES OF
ACTIVE VOLCANOES AND
HOT SPRINGS. THE SHADED
» PORTION 1S THE AREA WHICH
ACCORDING TO C.PSMITH WAS
COVERED WITH MATERIAL
EUECTED FROM MT TARAWERA
/N 1886.» HOT SPRINGS INDICATED ©\
SCALE (ee
Fic. 1.—Map of North Island, New Zealand, showing the zones of active
volcanoes.
calls the “‘Maori” line, since the Maoris probably migrated in a
general direction along that line. The ‘‘Samoan” and “ Hawaiian”
tR. Speight, ‘‘Geology,” Report on a Bot. Sur. of the Tongariro National Park
(Dept. of Lands, N. Z., 1908), p. 9.
THE ACTIVE VOLCANOES OF NEW ZEALAND 695
lines are supposed to cross the ‘‘Maori’’ line at their respective
points of greatest volcanic activity. Running practically parallel
to the fissured zone mentioned above, there is another zone con-
taining numerous extinct or dormant volcanoes stretching along
the eastern border of the island from the great Mount Egmont
through the Auckland district, where over sixty craters, mostly of
small magnitude, appear. There seem to have been, also, another
line of disturbance and a great fault running from the north-central
176 25" 176°30'
TARAWE RA
fe
> MT TARAWERA
AKE CRATER
SS
ee WA
: Proton Q
who
OKAKO
LAKE ‘
ae
MAP
SHOWIVG THE GREAT FISSURE IN MT TARAWERA,
THE PRESENT ROTOMAHANA LAKE (OUTLINED THUS—~)
THE LAKE BEFORE THE ERUPTION (OUTLINED THUS ss)
AND THE SMALL LAKES /N EXISTANCE JUST AFTER
THE ERUPTION (QUTLINED THUS =)
SCALE
ay D,
Lt te MOT LAKE LAK
snes
Fic. 2.—Map showing the course of the great fissure
part of the island nearly northwest to Hauraki Gulf and passing
through the Waihi mining district. The prominent scarp of this
fault may be seen from Morrinsville Junction in going from. Auck-
land to Rotorua, and it is necessary to ascend this steep slope
to reach the plateau before arriving at Rotorua. The streams
descend rapidly over this scarp, which is a prominent physiographic
feature of the landscape. This graben fault has aided in producing
the lowland stretching from Hauraki Gulf toward the central
portion of the island.
606 E. S. MOORE
These major zones of disturbance run parallel to the main
structural features of the islands resulting from the orogenic and
epirogenic movements which produced New Zealand and the
adjacent islands.
HISTORY OF THE LATER VOLCANIC ACTIVITY OF NEW ZEALAND
' There seems to be a general agreement among New Zealand
geologists that there were extensive post-Jurassic and pre-Mioeene
movements, resulting in much folding and in bringing the islands
nearly to their present geographical condition. The rocks formed
up to this time indicate, according to Marshall,' that the present
islands occupied a zone along the border of a continent now lost to
sight. The folding raised the mountain ranges from the sea bottom
and determined the major structures of the islands. ‘There have
been numerous oscillations since that time, but these have not
materially altered the main structural features. Following these
great disturbances, which may be correlated with those of America
and Europe, there was inaugurated an important stage of igneous
activity which became very prominent in the Miocene and has
continued, more or less actively, since that time. There was some
igneous activity during the Jurassic, and even then hypersthene-
andesites, so common in later periods, began to make their appear-
ance. Igneous rocks of this age are found, according to Park,? in
the Hauraki Peninsula, while the andesites and rhyolites of the
Canterbury district in South Island are regarded by some geologists
as Jurassic.
The greatest period of activity seems to have opened in the
Middle or Lower Miocene and to have extended dnto the Pliocene,
and even into the Recent, in North Island. During the Miocene,
which was also characterized by extensive orogenic and epirogenic
movements, the main centers of activity were the Otago, Banks,
and Hauraki peninsulas. The rocks of these areas generally rest
on Omaru sediments, which are regarded as Early Miocene. In
the Otago Peninsula the alkaline rocks were erupted at this time;
in the Banks Peninsula, rhyolite, andesite, and basalt; and in the
*P. Marshall, Geology of New Zealand (Wellington, N. Z., 1912), p. 188.
2 James Park, The Geology of New Zealand (Whitcombe & Tombs), p. 82.
‘THE ACTIVE VOLCANOES OF NEW ZEALAND 697
Hauraki Peninsula, andesites followed by rhyolites. Probably
the andesites extending north of Auckland up to North Cape were
contemporaneous with those mentioned. The important gold
veins of the Waihi mines are connected with this period of eruption
as a later phase of the activity.
The great volcanic plateau occupying the central portion of
North Island consists largely of rhyolite and pumice with the later
extrusions of andesite and related rocks breaking through the rhy-
olites. The first evidence of the activity which produced the
plateau is found in the rhyolite gravels of the Miocene, but the
main eruptions are believed to be of Pliocene age because much of
the pumice is found resting on early Pliocene strata and some is
interbedded with them. The earliest igneous rocks of this plateau
are, therefore, rhyolite and the latest andesite. As to the source
of these acid rocks, there are factors which point to the Taupo
area as the center of the eruption. While the writer did not have
the opportunity of visiting Lake Taupo, he is convinced, after
visiting other lakes in this region and reading descriptions of the
Taupo basin, that these larger lakes in the central portion of the
island are old craters modified by faulting. There is so much in
common between such depressions and many of those of crater origin
in the Hawaiian Islands that their origin can scarcely be in doubt.
The early andesite eruptions of Ruapehu, Tongariro, Egmont,
Edgecombe, and related volcanoes occurred in the Pliocene, while
the basanites of the Auckland area are probably of Pleistocene age.
PETROGRAPHICAL PROVINCES IN NEW ZEALAND
There is such a close relationship between the rocks of the
Ruapehu-White Island and Egmont-Auckland zones that they may
be justly regarded as belonging to one province. The rocks of
Mount Egmont consist of hornblende-andesite and hornblende-
augite-andesite; those of the Auckland region of basanite, poor
in nepheline and probably lacking in this mineral in some cases;
those of Ruapehu of augite-hypersthene-andesite; and those of
White Island of hypersthene-andesite.
On the Coromandel Peninsula there were first eruptions of
andesites of various types followed by rhyolite and these again by
6
608 E. S. MOORE
hypersthene-andesite. It is probable that the great rhyolite
extrusions of the central plateau were contemporaneous with those
of the Coromandel Peninsula, and that the early andesite extrusions
of this region did not occur in the plateau area. There are dacites
in both areas.
Park considers that there are two other petrographic provinces
in New Zealand of late Miocene or early Pliocene age, these being
found on the Otago and Banks peninsulas.* In the former penin-
sula the rocks consist of an earlier series of phonolite, dolerite,
trachydolerite, andesite, basalt, and basanite; and a later series,
erupted on the eroded surface of the first, consisting of basalt with
probably andesite and phonolite. Cutting the lavas of the first
series are dikes of nephelite-syenite, augite-dolerite, and tinguaite.
Professor Marshall, who has made a detailed study of this area,
states that no regular order of eruption and no definite system of
differentiation in these various rocks have, so far, been recognized.
On Banks Peninsula there was a period of rhyolite eruption —
followed, after considerable erosion, by andesites and basalts.
From the evidence presented there does not seem to be any
regular order of eruption followed by rocks of the various types,
except that in practically all cases there is a tendency for the volcan-
ism to cease with the eruption of intermediate rocks, as andesites.
RUAPEHU, NGAURUHOE, AND TONGARIRO
These three large volcanoes are located near the center of North
Island at the southern end of the rhyolite plateau. Their craters
lie along a direct line, within a distance of less than fifteen miles,
and if this line be projected northeastward it will pass also through
Pihanga and Tauhara, volcanoes now extinct; then through
Tarawera, Edgecombe, and White Island. Ngauruhoe is situated
between the other two and almost on the side of Tongariro, in
such a way as to indicate that it has arisen in the later stages of
Tongariro as a subsidiary cone to this great volcano.
The rocks of all three of these volcanoes are similar, and consist
of augite-andesite with augite-hypersthene-andesite. The early
activity produced extensive flows of these rocks followed by
Park, op. cit., p. 147.
THE ACTIVE VOLCANOES OF NEW ZEALAND 699
alternating lava flows and fragmental deposits of the same material.
Ruapehu has not been in active eruption since early in the Recent
period, but Ngauruhoe and Tongariro continue to suffer regularly
weak outbursts. Evidence of this may be seen in Fig.6. Accord-
ing to Marshall there has not been a flow of lava from a New
Zealand volcano in historic times, but Park and Speight believe that
in 1869 a lava flow escaped from the northwest side of Ngauruhoe
and that the fresh appearance of this lava attests its recent origin.
Fic. 3.—Ruapehu (0,175 feet) from the Waiouru-Tokaanu road eight miles
distant. Looking across the Onetapu Desert covered with volcanic sand and cinders.
Ruapehu.—This is an enormous mass of red to dark-gray lava
and scoriae rising from a plateau region. Its height has been
placed by various writers at 8,878 feet to 9,175 feet above sea-level,
and the latter may be considered as the more correct figure. It
has a large crater, approximately a mile in diameter, cut into on
the south-southeast side by a great ravine, so that the rim of the
crater consists of a series of prominent peaks. The crater is
occupied by a glacier which surrounds a small, hot lake said to be
about 600 feet in diameter. According to various reports, the
water sometimes boils, and apparently it is the sulphur water from
this lake which issues from the northeast side of the mountain.
700 E. S. MOORE
The writer was unable to reach the lake on the date of his visit in
October, 1914, owing to the steepness of the ice walls between the
point where he reached the crater and the location of the lake, and
from the brink of the crater no sign of it could be seen in the snow
field within the crater.
The sides of the cone are covered with masses of andesite from
the disintegrated lava flows and with fragments of large bombs.
In some cases columnar structure is well developed in these flows.
Fic. 4.—Bread-crust structure in a portion of a large bomb near the foot of
Ruapehu.
In a fragment of a large bomb lying near the foot of the mountain
and almost buried in the snow and cinders, an excellent example of
bread-crust structure was found (Fig. 4). Small glaciers hang on
the cone, extending, in some cases, as low as 2,000 feet below the
crater rim.
Ngauruhoe.—This is a beautiful and symmetrical cone resting
on an upland base which was probably largely developed by
Tongariro before Ngauruhoe was of much importance. The
elevation of this mountain is placed at 7,481 feet by S. P. Smith,
and at 7,515 feet by Marshall. It is made up of a base of andesite
flows on which rests the cone, consisting of alternating lava flows
THE ACTIVE VOLCANOES OF NEW ZEALAND 701
and beds of tuff and agglomerate, with bowlders up to ten feet in
diameter. Some interesting examples of flows which appear to
have split, passing above and below beds of agglomerate and tuff,
Fic. 5.—Ngauruhoe from a point near the foot of Ruapehu. This view shows
how the cone is built on an upland largely developed by Tongariro.
Fic. 6.—Ngauruhoe (7,515 feet) showing the usual cloud of steam and sulphur
fumes rising from the crater.
702 E. S. MOORE
may be seen on the east side of the cone (Fig. 7). These are found,
on close examination, to be due to the viscous lava piling up and
becoming brecciated in movement, so as to resemble a bed of tuff
and agglomerate.
There was considerable snow and ice on the mountain when
the writer visited it in the spring season, early in October, but this
disappears in the summer and no glaciers remain here, as on
Ruapehu.
Fic. 7.—Apparent splitting of lava flows. This seems to be due to the viscous
lava becoming brecciated in movement so that it resembles tuff and agglomerate.
The liquid lava then flows over the fractured layer.
The crater may be entered on the north side, where the rim is
broken away and it is comparatively level on the bottom except
for two mud volcanoes on the floor and a deep depression on the
west side, the depth of which cannot be estimated since it is always
full of fumes. The diameter of the main crater is about 500 feet
and the height of the perpendicular walls on three sides of it was
estimated at 200 feet in the higher portions. In the small crater
there is a great deal of activity. Large clouds of steam mingled
with sulphur dioxide rise continuously, and at times detonations
like the crack of heavy rifle-fire may be heard. Considerable dust
is intermittently shot up from this crater and, as seen from Fig. 8,
THE ACTIVE VOLCANOES OF NEW ZEALAND 703
these explosions are occasionally quite violent. The explosion
which threw out the cloud seen in the photograph, and which
occurred on October 3, 1914, was said by some of the residents
of Waiouru, twenty-five miles distant, who witnessed it, to be one
of the strongest outbursts observed for at least two years. Up to
the time this occurred, on the date mentioned, no sign of activity
was seen around the mountain top, until at 9:30 A.m. this cloud
was suddenly shot up about 1,000 feet before being drifted away
Fic. 8.—Cloud of dust and steam blown from Ngauruhoe, October 3, 1914.
This explosion was much more violent than usual.
by a terrific wind, which was blowing at the time and prevented the
cloud from rising to a great height.
Fumaroles occur around the steep walls of the main crater and
well down the north side of the cone. On the northeast side of the
cone was seen some reddish, highly vesicular, ropy lava, which has
every appearance of being quite recent. As mentioned above, it
has been stated by a number of writers that this stream was
erupted in the year 1869, but not all New Zealand geologists are
in accord on this subject.
Tongariro.—There are many features which make it appear
that Ruapehu and Tongariro are major volcanoes with Ngauruhoe
704 E. S. MOORE
subsidiary to the latter. Tongariro is an immense volcano, but
with a cone of much less altitude than that of either of the others
just described. The history of this mountain has been very
Fic. 9.—Looking from Ngauruhoe into the center of the crater of Tongariro and
showing the Red Crater in the foreground.
Fic. 1o.—One of the later flows of andesite from Ngauruhoe
THE ACTIVE VOLCANOES OF NEW ZEALAND 705
similar to that of Ruapehu. According to all the descriptions
given, the lavas are andesites, mostly of the augite-hypersthene
type, with small amounts of the hornblende-hypersthene type in
some of the earlier flows. The cone consists of alternating lava
flows and beds of scoriae. According to Speight! the height of the
cone was greatly reduced by a terrific explosion, which was followed
by extensive lava flows, and which blew 2,000 to 3,000 feet off the
mountain. The crater rim now has a maximum altitude of 6,458
feet and is made up of a number of peaks surrounding several minor
craters. One of these, known as the Red Crater because of the
red color of the lava, is situated near the center of the main crater.
Another, called Te Mari, lies on the northeast corner, and a third
known as Tama is southeast of Ngauruhoe. Tama is believed to
be part of the old crater rim, even if it lies beyond Ngauruhoe, and
this is proof of the subsidiary character of the latter crater. All
these craters are in the solfataric stage, but Te Mari is said by
Speight to suffer explosive activity at times and to throw out ashes
and stones. It was from this crater that the flow of andesite
poured down through the forest on the flank of the mountain, and
the conditions indicate that this eruption occurred at a compara-
tively recent date, although not within historic time.
In a depression on the main crater floor there is a small lake,
called Blue Lake, lying at an elevation of about 5,500 feet. This
lake, Te Mari, Red Crater, and Ngauruhoe all lie in almost a
straight line, and they are apparently located on a fissure, or narrow
zone of weakness, in the earth’s crust. The Ketetahi Hot Springs
are situated a little to the east of the line mentioned and well down
on the northern flank of the mountain. They exhibit very strong
thermal activity. Lying between Ngauruhoe and Ruapehu there
are two small lakes, which probably owe their origin to some of the
_ explosive activities of Tongariro.
MOUNT TARAWERA
Much has been written on Tarawera but many of the original
works are out of print and unavailable. Reports have been
prepared for the government bureaus of New Zealand by A. P. W.
i Speight, of. cit., p: 11.
706 E. S. MOORE
Thomas, Sir James Hector, and S. P. Smith, while other descrip-
tions may be found in the works of Hutton,' Marshall,? and Park.s
The special interest in this volcano lies in the great eruption of
1886, which produced results of much scientific, economic, and
humanistic importance. The opening of the yawning chasm
through the mountain, followed by the distribution of ashes over
thousands of square miles of country with the accompanying
destruction of life and property, is a matter of interest to every
traveler who approaches this region.
Mount Tarawera was a small, nearly flat-topped mountain of
rhyolite about 3,600 feet high and approximately 2,500 feet above
Lake Tarawera lying at its base. There are on the mountain three
prominences, known as Wahanga, Ruawahia, and Tarawera, the
latter giving its name to the mountain as a whole. The structure
was that of almost horizontal beds of pumice and flows of rhyolite,
which had been poured out of some adjacent volcano or fissure, and
which made up part of the rhyolite plateau in the central portion
of North Island. Up to the time of the great explosion there was no
evidence of a crater in the mountain, but it is situated in the zone
of fissuring which runs from Ruapehu to White Island, and previous
to the eruption there were numerous hot springs and geysers in the
vicinity of the present Lake Rotomahana. It is close to Lake
Tarawera, which has every appearance of being an old crater modi-
fied by local subsidence. The walls are steep and the water near
the shore is deep in many places. The same condition exists in
Lake Taupo and it may be concluded that all the steep-walled
lakes in this region are of crater origin. The whole region lying
between Tarawera and Rotorua is perforated with craters, hot
springs, and geysers.
THE ERUPTION OF 1886
During the night of June 10, 1886, violent rumblings were heard
and minor earthquakes experienced in the region surrounding the
mountain. These increased in violence until about 2:00 A.M., when
tF. W. Hutton, Report on the Tarawera Volcanic District, Wellington, 1887.
Also ‘‘The Eruption of Mount Tarawera,” Quar. Jour. Geol. Soc., XLII, 1887.
2 Marshall, op. cit., p. 107. 3 Park, op. cit., p. 166.
THE ACTIVE VOLCANOES OF NEW ZEALAND 707
the main eruption commenced and the great fissure began to open
in the mountain, commencing at the north end in the hump called
Wahanga. It passed through Ruawahia toward the basin now
occupied by Lake Rotomahana and formerly containing the small
lakes, Rotomahana and Rotomakariri. It opened under the lakes
about 2:30 A.M. with a terrific roar and a cloud of steam which rose
over 15,000 feet high. This no doubt was due to the water rushing
into the heated zone and producing a great explosion of steam.
The result was the opening of a large pit, now occupied by Lake
Rotomahana, while the débris was scattered widely over the
country. The finer materials were'carried out to sea over the Bay
of Plenty, as indicated on the accompanying map (Fig. 1). It
has been estimated that from the great fissure from 520,000,000
to 620,000,000 cubic yards of material was thrown out and this was
spread over an area of almost 6,000 square miles, of which 1,500
square miles were damaged more or less severely from the agri-
cultural standpoint. All habitations within. four miles of the
mountain were destroyed and 116 people killed. Most of these
were natives, and while the majority of them were killed by the
falling materials burying them, some around Rotomahana, where
the natives often gathered, were literally carried away by the
explosion. At Te Wairoa the buildings were crushed in and all
vegetation destroyed or very severely damaged. There is still
very little vegetation near the mountain, but it is interesting to see
how quickly it has re-established itself at Te Wairoa, where the
eucalypti are already fourteen to fifteen inches in diameter and
other trees of less rapid growth are eight inches. The fern, like
the braken of this country, establishes itself very quickly and
flourishes on the acid soil. In many places the charred remains of
trees are found, not only in the ashes of this eruption, but also in
the ashes of earlier date.
The main eruption lasted about five hours, although the more
violent part was probably over in less than an hour. Earthquakes
continued for many days and there seems to have been some unusual
activity in the hot springs around Rotorua. There have also
t According to S. P. Smith the measured height was 15,400 feet. ‘‘The Eruption
of Tarawera,” A Report to the Surveyor-General, New Zealand, 1887.
708 E, S. MOORE
been reports of sympathetic action in Ruapehu, White Island, and
other places along the volcanic zone.
Previous to the eruption of Mount Tarawera there were numer-
ous hot springs and geysers in the area occupied by the present
Lake Rotomahana, and the famous Pink and White sinter terraces
were situated well within the border of the present lake.
THE GREAT FISSURE
As stated above, the eruption of Mount Tarawera began at the
northern end of the mountain and progressed southward with the
opening of an enormous fissure. This chasm is about 82 miles
long, 1% miles wide in Lake Rotomahana, and goo feet deep in the
mountain. It is one of the most extraordinary openings to be
found anywhere in the earth’s crust (Fig. 11). Where it cuts
through the mountain it takes the form of several deep, narrow
craters in linear succession, separated by wedges of rock not blown
out by the great explosions. The deepest opening is about goo feet
and it is about 1,000 feet wide at this point. In some places the
crater walls are nearly vertical, but in others they have a gentle
slope and can be descended to the bottom. ‘There are a few small
fumaroles, but they are no longer important. Along the brink
of the chasm there is about 175 feet of highly colored, red and
variegated scoriae deposited on top of the rhyolite materials thrown
cut of the fissure, but there is no evidence of a lava flow. ;
The fissure runs down the mountain side and through Lake
Rotomahana, where it is 520 feet deep and has very steep walls.
in some places. It reached its maximum width here, where it is
14 to 14 miles wide. The present lake is about 4 miles long and
2 miles wide, and it covers the areas formerly occupied by the old
Lake Rotomahana and Lake Rotomakariri. In the fissure,
immediately after the eruption, there was a small lake called Hot
Lake, but gradually the whole depression became filled with water.
The explosion completely destroyed the Pink and White sinter
terraces, which were located within the present basin rim, and
fragments of them may be picked up for miles around where they
are mingled with the other ejectementa from the fissure. There is
still much thermal activity around Rotomahana, the name of which
799
THE ACTIVE VOLCANOES OF NEW ZEALAND
aNOAYI IO} YSI] OY} YIM oANssy oy} JO YULIq oY} Suoje uses ApOurjsIp st seIIOIS po Jo JaAe] oy} GJ pur PF UT
0} aInssy UMOP SuTyoo] ‘9
fuorjiod yeijue9 dsap ‘g
‘
pua yyou 9y} ‘Pv
“RIOMEILT, JUNOF YSNoIYy} sinssy yvor
Q
ne)
‘WI SutAyiepun
“BURYLUIOJOY IVT
syL—II ‘Oly
710 E. S. MOORE
p)
signifies ‘‘warm lake,” and steam rises from many parts of the
shore, especially near the northwest corner where the terraces and
other hot-spring phenomena were most prominent before the
eruption. The color of the water is a sort of dirty, greenish gray,
like that of glacial streams, this hue being caused, no doubt, by the
large amount of extremely fine particles of mineral matter suspended
in the water.
Continuing westward the fissure passes through Black, Inferno,
Echo, and Southern craters, all of which exhibit considerable
Fic. 12.—Lake Rotomahana, through which the great fissure passes from end to
end. Looking westward from Mount Tarawera.
thermal activity at the present day. The basin of the extraordinary
Waimangu Geyser, now inactive, is located on this line a short
distance from Lake Rotomahana. This geyser became active
in 1900 and continued more or less irregularly until 1908, when it
ceased to act. It has been reported by various reputable author-
ities that it often threw water and mud to a height of from 1,200
to 1,500 feet. With the extinction of this geyser the surrounding
springs became more active. The Waimangu “blow hole,”
situated southwest of the geyser orifice, blows hot water and steam
for two minutes and is then quiescent for seven. In Echo Crater
THE ACTIVE VOLCANOES OF NEW ZEALAND 711
the floor and walls are dotted with hot springs and fumaroles, and
around some of these springs a great deal of iron pyrite is being
deposited on pebbles, particularly in a spring called “The Frying
Pan.” The pyrite becomes disseminated in the sinter and to some
extent it impregnates the thermally altered rhyolite. It seems to
owe its origin to the reaction between H.S and some iron salt, which
in all probability is the chloride. ‘The sulphide coats the pebbles
with a black, smooth, waterworn layer which later tends to assume
more nearly the appearance of typical pyrite. An assay was run
on this pyrite deposit to determine the presence or absence of gold,
and no trace of gold or silver was found. It seems probable that
the pyrite in the sinters around Rotorua is of the same origin, and
the large deposits of sulphur around the springs near Lake Rotorua
appear to be due to the oxidation in the air of the H.S so plentiful
in these waters.
While the great fissure practically ends at the Southern Crater
there are some smaller fissures and faults in Earthquake Flats
which indicate the extension of the disturbance beyond the main
fissure. There are lines of former movement which were again
depressed a few feet.
DETAILED DESCRIPTION OF THE ROCKS OF MOUNT TARAWERA
This mountain was originally made up of interbedded rhyolite
and rhyolite pumice, with streaks of dark gray to black, spherulitic
obsidian running through the rhyolite. The bands often have the
appearance of irregular dikes in the rhyolite, but they are probably
due to the varying rate of cooling in different parts of the flows.
The dark obsidian contains many fragments of the lighter rhyolite,
and in some cases these have the appearance of being partly
absorbed. This may be explained by the rhyolite fracturing on
the cooled surface, permitting the liquid beneath to pour out
around the brecciated fragments and to cool quickly. Good
examples of this spherulitic obsidian were found on the road
leading from Te Wairoa down to the landing on Lake Tarawera.
Fragments may also be picked up among the débris from
Tarawera, showing that the rock exists in the deeper beds in the
mountain.
712 E. S. MOORE
The rhyolite is quite fresh, brittle, and friable. Thin sections
show that it contains a very deep dark-brown biotite, some augite,
and, in one case, a grain of hypersthene, in addition to orthoclase,
albite, and quartz which is very Blassy and brilliant. The ground-
mass is usually mostly glass.
The obsidian consists of a brittle, dense, black glass, showing
flow structure. It is full of spherulites from 0.05 mm. to 3.5 mm.
in diameter. The glass contains also phenocrysts of green horn-
blende, orthoclase, and zonally built crystals of orthoclase and
albite. The smaller spherulites consist of radiating needles of
feldspar, while the larger ones are nearly solid glass around the
center, with radiating dark lines and with concentric spheres
io OS
ce a re
F1c. 13.—Bombs of andesite and basalt from Mount Tarawera. In two of them
the light-gray cores of rhyolite may be seen (g natural size).
becoming more distinct toward the exterior. These spheres are
alternately brown and gray. The outer thick zone is brown and
shows only a glass without crystal structure. -The other zones show
radiating small crystals of feldspar under the high-power micro-
scope, but there is so little crystal structure that only a very slight
darkening and brightening can be observed on rotating the section
between crossed nicols, and there is almost no difference in bire-
fringence between the spherulite and the surrounding glass. A
distinct bending of the microlites in the glass around small spheru-
lites may often be observed.
Bombs.—During the eruption of 1886 a considerable amount
of intermediate to basic rock was ejected from the crater. It has
been estimated that from 520,000,000 to 620,000,000 cubic yards
of material was blown out of the great fissure. This was largely
THE ACTIVE VOLCANOES OF NEW ZEALAND 733
rhyolite, but about 175 feet of dark, reddish-brown scoriae con-
sisting of ashes, lapilli, and bowlders of vesicular and ropy lava
lies along the brink of the great chasm through Mount Tarawera.
The lava, which was the last to fall on the mountain, except some
material from the basin of Rotomahana, welled up beneath the
chasm and was caught in the big explosion. It was blown to
fragments and thin layers of the fine material are mixed with the
lighter colored tuff from the rhyolites around Rotomahana. It
formed irregular masses of scoriaceous and ropy lava up to two
feet in length, while it quite frequently formed spherical and oval
bombs (Fig. 13). These bombs occur in great numbers around the
foot of the mountain. ‘The most peculiar are those with a core of
rhyolite and an enveloping coat of andesite or basalt. They owe
their origin to the fact that fragments of rhyolite were engulfed
in the more basic lava, and when the explosion occurred these were
hurled into the atmosphere with a rotary motion so that the viscous
molten material became well wrapped around the core of solid rock.
As a rule, this core is not exposed until the bomb is broken open.
They all show the bread-crust structure well developed owing to
the shrinking of the cooling, molten coat around the solid interior.
In the specimens examined there is a sharp line of contact between
the two rocks and there is no evidence of fusion of the rhyolite.
An examination of thin sections of the more basic rock showed
in one case much dark, grayish-brown, vesicular glass containing
numerous little laths of feldspar, a little augite, and a few small
phenocrysts of enstatite. In another specimen the same minerals
were found, with the exception of augite. In one small bomb the
feldspars were identified from their extinction angles as anorthite
and bytownite, and this same specimen contained traces of quartz,
possibly due to absorption of some of the acid rhyolitic material
before ejection. It was carefully examined for nephelite, owing to
the reported occurrence of nephelite in the Auckland lavas, but it was
found to be optically positive and to lack any sign of cleavage. Very
small crystals of augite were present. The rock is a quartz basalt.
Dr. Marshall mentions hypersthene-augite-andesite in the
bombs from Mount Tarawera,’ but no hypersthene has been found
t Marshall, op. cit., p. 102.
714 E. S. MOORE
by the writer, the orthorhombic pyroxene in all cases being like
enstatite.
From the description given it is evident that these rocks vary
from andesite to basalt and that they represent a much more basic
phase than any rocks previously erupted in the vicinity of Mount
Tarawera. ‘The sequence is very similar to that in all the other
volcanoes in this petrographic province.
\
GLACIATION IN THE VOLCANIC ZONE
There has been much discussion in New Zealand in recent years
regarding glaciation in North Island. Outside of the comparatively
small glaciers on Ruapehu the writer did not see any evidence of
glaciation. Around both Ngauruhoe and Ruapehu there were
many bowlders which had grooves very similar to those often made
by glaciers. It was surprising to find, however, that in practically
all cases these were not due to glaciation, but probably to the
action of one mass of rock falling on another when hurled from the
craters. This was proved by the fact that the groove would often
end abruptly against the wall in a re-entrant angle in such a way
that it could not have been produced by glacial action.
FOOTHILLS STRUCTURE IN NORTHERN COLORADO
VICTOR ZIEGLER
Colorado School of Mines, Golden, Colo.
OUTLINE
INTRODUCTION ie
STRUCTURAL FEATURES
The Foothills Monocline
Supposed Local Unconformities
The Golden Area
Structural features at Golden
Eldridge’s ‘‘ Arch Hypothesis”’
Comments on the Arch Hypothesis
Proposed Alternative Explanation
The Boulder Area
Minor Structural Features
Minor Folds
Minor Faults
SUMMARY
The Master-Monocline
The Main Faults
Conclusion
INTRODUCTION
The area under discussion comprises a part of the foothills of the
“Front Range” of Colorado, west and north of Denver, and extends
from Morrison about 70 miles north to Fort Collins. No attempt
will be made to discuss the stratigraphy of this district nor to
describe any of the geological formations in detail; only the
structural features will be emphasized.
STRUCTURAL FEATURES
THE FOOTHILLS MONOCLINE
The “‘master structural feature’”’ of the foothills has been well
described by Fenneman’ in his discussion of the geology of the
Boulder district, and also by Eldridge in the monograph on the
U.S. Geol. Survey Bulletin 265, 1905, pp. 41-43. :
715
716 VICTOR ZIEGLER
geology of the Denver basin.t The following description is taken
from the latter publication:
The normal appearance of the foothills is that of a mountain mass of
Archean rocks, fringed at an average distance of one half or three quarters
of a mile by a sharp serrated ridge of Dakota sandstone, the valley between the
two being occupied by the formations of the Trias and Jura. Above the
Dakota come .... the Benton, the Niobrara—this generally constituting
a second smaller reef outside the Dakota—the Pierre, the Fox Hills, and the
Laramie, the basal sandstones of the Laramie again forming either a low roll
in the ground or an actual comb of rock slightly projecting above the surface
of the surrounding prairie. To the east of the Laramie... . appears in the
southern portion of the area yet another comb formed by the conglomerates
at the base of the Arapahoe series. Finally this is followed by... . the
Denver formations.
To this description it is well to add that as a general rule the
Lyons formation forms a prominent, though low, hogback in the
strike valley to the west of the Dakota, and that in the northern
part of the foothills the Archean-Fountain contact forms a promi-
nent strike valley, and that here the Fountain is characteristically
developed into a high, precipitous hogback, usually capped on its
crest and dip slopes by the Lyons sandstone. Here also the
Arapahoe and Denver formations are absent, and the Laramie
lies almost horizontal twenty miles or more to the east of the
foothills. In the southern part of the area the dips in the Foun-
tain, Lyons, and Lykens average from 35° to 50. ‘These increase
gradually eastward until they become vertical or even overturned
in the Laramie, dips as low as 75° west being noted. Farther
eastward these flatten within a few hundred feet from vertical
into practically horizontal in the upper part of the Arapahoe or
the base of the Denver.
No such variation is noted in the northern part of the area
under discussion. West of Loveland dips of 40° are rare and
occur only locally as the result of special conditions. The steepest
dips occur here in the Dakota or Morrison. It is also worth noting
that in the southern part of the area, where steep and overturned
dips are met with, the formations have turned practically hori-
zontal within two miles of the Archean contact—in one extreme
U.S. Geol. Survey, Monographs, XXVII, 1806.
FOOTHILLS STRUCTURE IN NORTHERN COLORADO 717
case (at Golden) even within 4,500 feet from the contact; while
in the northern part of the area dips of 15 ° east are found as much
as 10 miles east of the Archean sedimentary contact.
These contrasts in dips indicate that the foothills fold differs
in shape and intensity over this area. The generally low dips
near Fort Collins and Loveland indicate an ideal, fairly gentle,
monoclinal fold, while the steep and overturned dips to the south
indicate an S-shaped fold of pronounced type. The fold proper is
well described by Fenneman’ as follows:
The master structural feature of this region is the great upturn of‘the
strata against the mountain range. ... . The first Archean belt west of the
foothills is a dissected plateau . . . . from 6,500 to 7,000 feet above sea-level at
its eastern edge, where it ends abruptly and is flanked by the Fountain sand-
Stone.) 5.1. The height . . . . above the plains is nearly 1,000 feet... . .
Five miles east of the base of the foothills the Archean surface is at least 9,500
feet below the surface, or 4,200 feet below the level of the sea. The real face
of the granite plateau is therefore about 2 miles high, and this enormous rise
is accomplished in 6 miles.
The opinion is also expressed that the greater part of the
monoclinal flexure is the result of subsidence during the deposition
of the various sedimentary formations, a conclusion passed on the
assumption that the ancient shore line was located along the line of
the present Front Range.
This generalization was, however, based on a small area of the
foothills in which the characteristic structure is not well developed.
Further, Lee? has proved since that all the Cretaceous formations
up to and including the Laramie formerly were continuous over the
present site of the Front Range. Consequently their present atti-
tude must be due to orogenic forces, and the statement that ‘‘the
process of mountain making gave to the granite plateau west of the
foothills a comparatively small relative uplift above the plains”’ is
erroneous.
A careful study of the dips and strikes of the formation south of
the Boulder area and the reconstruction of a fold based on these
dips as well as the occurrence of the lower formations as ‘‘inliers”’
1 Op. cit., pp. 41, 42.
2 U.S. Geol. Survey, P.P. 95-C.
718 VICTOR ZIEGLER
Lararme
LOVELAND Prerre 8 fox Hills
Fountain & fon
___ Lararnie
ALTONA Prerre & Fox Hills
Co/orado
ee
Prerre & Fox Hills
Fountain & Lyons
DRY CREEK
Prerre & Fox Hills
BOULDER
Colorado
Harrison & Dak
Fic. 1a.—Reconstructions of the original monoclinal fold of the foothills of the
Colorado Front Range.
FOOTHILLS STRUCTURE IN NORTHERN COLORADO 719
Pierre & Fox Hills
COAL CREEK
Colorado
GOL DEN
MORRISON
Fic. 1b.—Reconstructions of the original monoclinal fold of the foothills of the
Colorado Front Range.
720 VICTOR ZIEGLER
in the Archean some distance west of the foothills, as for example
west of Colorado Springs and also west of Loveland, furnishes
good reasons for the belief that the formations underlying the
Dakota also extended over the Archean, or at least for a considerable
distance westward. In areas undisturbed by special local condi-
tions, the Fountain, Lyons, Lykens, and Morrison show as close an
agreement in dips and strikes with the overlying formations as
would be expected for conformable rock series; and all show
beautiful parallelism in folding with the younger rocks. This
would be a rather peculiar coincidence, not to be expected except
in rocks simultaneously folded.
The accompanying diagrams show, drawn approximately to
scale, reconstructions of the original monoclinal fold worked out
from dips and strikes at the localities noted. In each case faults
and other local irregularities which would unnecessarily complicate
the fold have not been indicated. ‘Their possible effect, due to dis-
placements of parts of the fold and consequent introduction of
anomalous dips and strikes, has been carefully considered in each
case, and, where necessary, corrections have been made in the
flexure of the monocline. These reconstructions are merely in-
tended to show the variation in the shape and curvature of the
foothills monocline in its major outlines (Fig. 1, a, b, pp. 718-19).
It is interesting to note that in older interpretations of the
monoclinal fold in this region the overturn is located underneath
the surface and not above the surface, as the present writer has
drawn it. Considering no more than the present dip and strike
relationship along an east-west line at such localities as Golden
or Boulder, and considering these to be located on the same hori-
zontal plane on the monoclinal fold, no other logical interpretation
is possible. If, however, the variations in dip be carefully plotted
along the axis of the fold, and if we carry in mind the fact that the
dips as observed at Golden and Boulder represent different hori-
zontal planes on the monocline due to the disappearance of thou-
sands of feet of strata, the type of overturn postulated by the
writer becomes necessary in order to explain observed conditions.
This will perhaps be clearer after a discussion of the structure at
Golden.
FOOTHILLS STRUCTURE IN NORTHERN COLORADO 721
LOCAL UNCONFORMITIES (SO-CALLED “ ARCHES”’)
Certain peculiarities in the distribution of the geological
formations along the foothills, especially well developed near Golden
and Boulder, have been explained as the result of unconformities
due to local arching in the Cretaceous sea.
The Golden area.—The sketch map of the vicinity of Golden is,
with a single correction, the map by Eldridge given in his report
on ‘The Geology of the Denver Basin.”* ‘The areal distribution of
the formation has been well traced in this map and has proved
essentially correct, while the general features of the area in
question (Fig. 2) have been well described in the following
words:
The topography shows a marked variation from that normal for the
foothills region in general. ... . For mile after mile along the mountains the
normal topographical features may be traced with unswerving regularity, but
within the area to be described they undergo rapid change, and .. . . in the
vicinity of the town of Golden they are lost to recognition. For a distance of
over a mile north of the town, and an equal distance south of it, the Dakota
hogbacks have completely disappeared; the low Niobrara ridges cease to
exist at a point about a mile north of Bear Creek, not to appear again until the
region of Van Bibber Creek, 10 miles to the north, is reached; the Laramie
sandstones with their coal have gradually approached to within 500 feet of
the Archean at Clear Creek, the variation in their strike from that of the
Triassic and Dakota outcrops below being apparent to the most casual observer.
. . .. The lines of stratification are delineated clearly upon the surface and °
display a distinct tendency to group themselves, with respect to direction,
into two well-marked assemblages—the one embracing the formations of the
Colorado and all below, and maintaining for the greater part of their extent
the same parallelism to the general trend of the foothills which they have held
beyond the affected area; the other embracing the Montana and younger
formations, and though maintaining a parallelism of strike within themselves,
nevertheless abutting against the older formations, in fact approaching the
range proper in a broad, well-marked, and regular inward-sweeping curve, the
center of its arch lying a short distance north of Clear Creek. The features
just noticed again occur, in a minor degree and in a manner not at first liable to
attract attention, in the relations between the Dakota and underlying beds
nearer the middle of the area, where the beds of the younger formation lie
across the edges of those of the older.
7 Op. cit., p. 83.
2 Op. cit. pp. 83-84.
722 VICTOR ZIEGLER
Eldridge discusses each geological formation in great detail,
and the care with which all structural features are described is
worthy of special note. Rather than cite at length from this
report the writer has arranged the pertinent data given in tabular
form."
STRUCTURAL FEATURES AT GOLDEN
Permo-Trias (Fountain, Lyons, Lykens)
1. Rapid disappearance of strata successively from the top down-
ward as they approach Golden. Lyons disappears 1 mile north
of Clear Creek, and }-mile south respectively. Lykens dis-
appears about two miles north and south of Clear Creek.
. Disappearance of Lykens is sudden north of Clear Creek; gradual
to the south.
3. Discrepancy of 10° in strike with upper formations on disturbed
area.
4. Normal thickness of Morrison is present where much of Lykens is
missing.
Morrison
t. Missing for a distance of about 13 miles.
2. Discrepancy in strike (10°-15°) with younger Dakota and older
rocks on disturbed area—especially marked in part north of
Clear Creek.
3. Sudden disappearance of Morrison north of Clear Creek; gradual
disappearance south.
Dakota and Purgatoire
1. Disappears from bottom up and top down.
2. Discrepancies in strike with rocks above and below on disturbed
area.
3. More sudden disappearance on north side of area.
4. Marked crumpling on north side—frequent changes in strike, not
shown by Laramie, which is only 600 feet to the east.
5. Normal dip 45°. Dips 90° and overturned over disturbed area.
6. Remarkable crumpling and recumbent folding south of Golden.
(Not noted by Eldridge.)
Benton and Niobrara
1. Completely disappear from top downward. Benton absent for
distance of about 5 miles; Niobrara for 9 miles.
2. Conformable in dip and strike to Dakota, but not to Montana on
disturbed area.
3. After disappearance of Niobrara, overlain by successively higher
Montana, beds as it approaches Golden.
iS)
1 Op. cit., pp. 91-97-
2
FOOTHILLS STRUCTURE IN NORTHERN COLORADO
JoAUIqG = apy (BIvIGOIN] puke UOJUIG) Ope10jOD = IM
soyedeiy= vy a10jeBINg pue vjoyeq ‘UOSIIIOP[ =U l-py
sUIeIeT= 7 suoyA'T pur ‘suod’'T ‘urezuno = mt
(I119Iq Pure S][IF{ XOq) eueyuoPT= wy uvsypIVY= UsLy
S}[NeVyJ MOYs soul, AAVIFY “UIploy Jo APUINIA ay} Jo deur yoJaYS—s “OT
ffi zl hg
|
MEA
_~*
VICTOR ZIEGLER
“I
bo
aS
Montana (Pierre, Fox Hills)
1. Pierre completely disappears from bottom upward. Absent for
distance of about 43 miles. Upper part of Fox Hills, only present
at Golden.
2. Conformable in dip and strike to Laramie. Shows a discrepancy”
of 15°—20° in strike with older rocks.
3. Steepest dips on area at Golden—go° and overturned (80° W).
Laramie :
1. Broad, sweeping curve by which it is gradually carried to the west-
ward until at Golden it lies within 4,500 feet of the Archean.
2. No thinning.
ELDRIDGE’S HYPOTHESIS OF AN ‘‘ ARCH” AT GOLDEN
The abnormal conditions tabulated above are all explained as the result of
a series of unconformities at the horizons where these occur.
There is postulated a headland of anticlinal structure with
axis perpendicular to the present trend of the foothills. This
has been named the “‘Golden Arch” (Fig. 3). No attempt will
be made to discuss the so-called “‘arch hypothesis”’ in detail. For
such, the reader is referred to previous publications." It will be
noticed that the lines along which certain formations disappear
are roughly parallel to the general strike of the foothills formations;
and that as we approach Golden the formation to the west of such
a line—that is, the older—disappears from the top down; while
the formation to the east—that is, the younger—disappears from
the bottom up. In the ‘“‘arch hypothesis” such disappearance is
explained as the result of the erosion of the older formations from
the top of the rising headland, followed by a subsidence and a
gradual overlapping of the younger formations against the sides
and eventually over the top of this arch. The older formations
are missing because of erosion; the younger, because of non-
deposition.
Comments on the Arch Hypothesis: A number of points
brought out in a former discussion as well as some additional data
from field work seem to show serious weakness in the arch hypothe-
sis. ‘These points are in part facts described in detail by Eldridge
and summarized above, in part facts discovered since, or at least
not specifically noted by him, and in part certain fundamental
U.S. Geol. Survey, Monographs (1896), X XVII, 91-97.
725
FOOTHILLS STRUCTURE IN NORTHERN COLORADO
‘IAUIqd = epy
soyedviy= vy
IWUVIVT= 71
SHH XO = 4f¥
wWdig= ¢¥
BIVIGOIN= UY
(dn w10}}0q wor) UMOYs are V pu ‘E ‘z
‘
I SA881S
uojuog= gy
(e110jvsIng pur) ejoyeq= py
uosHIOf[T= we
(Ssurm0A MM Joddy) sueyAT= mar
(SUIWMIOAM, JAMO'T) SUOA'T pu UTeJUnOT= yar
uvdypIy= us
‘(a3pUpl[A Joye) yore Uspjoy oy} Jo yUowdopaop oy} ul sesveIG—'f “oly
726 VICTOR ZIEGLER
assumptions that appear unsound or at least unwarranted to the
writer. The chief objections may be summarized as follows:
I. Objections based on facts:
1. The steepest dips occur where the formations disappear.
For what reason, if their disappearance be due to an unconform-
ity?
2. The greatest amount of overturning (60° W) is where the
greatest thickness of strata is missing. The same question may
be asked as for point No. 1.
3. Crumpling is common along the lines of the supposed uncon-
formities.. Why ?
4. Crumpling is relatively more severe on the axis of the sup-
posed arch, that is, on the line where the greatest thickness of rock
is missing. Why?
5. Crumpling is relatively more intense at those points where
the formations disappear suddenly. Why this coincidence if their
disappearance be due to erosion or non-deposition ?
6. There is a remarkable coincidence of an association of
maximum divergence of strike lines, steep and overturned dips,
maximum crumpling, and the sudden disappearance of a formation.
The simultaneous relationship of these to an unconformity is not
clear.
7. There is an absence of shore facies in the sedimentaries.
The existence of the Golden arch presupposes the existence of
shore conditions, but none of the formations supposedly deposited
against this arch show any change in lithological character upon
crossing or approaching it.
8. No similar structures have been recognized elsewhere in the
Rockies.
g. On the bluffs immediately to the south of Clear Creek the
Dakota shows a dip of 60° westward, while less than 100 feet west
the Fountain shows a dip of 40° eastward. These dips cannot be
explained on the basis of an unconformity. (Not noted by
Eldridge.)
t See also H. B. Patton, ‘“‘ Faults in the Dakota Formation at Golden, Colorado,”
Colo. School of Mines Bulletin, 111, No. 1 (1905), pp. 26-32. A complete overturn
through 180° is here described.
FOOTHILLS STRUCTURE IN NORTHERN COLORADO 727
10. There are many minor strike thrust faults clearly shown, as
well as dip faults with steep fault planes and a hingelike displace-
ment toward the east—all of which are most pronounced in the
immediate vicinity of Golden.
II. Objections based on interpretation:
t1. The arch hypothesis presupposes the existence of a shore
line immediately to the west. The work of Lee’ has proved the
former extension of the Cretaceous formations in a continuous
sheet across the entire area of the Front Range. Hence this whole
area must have been an epicontinental sea, and consequently a
“headland” similar to the arch could not have been present at
Golden.
12. Too much oscillation required of a small local area. Such
rapid alternations of up and down movements are a strain on
credulity.
13. It seems incredible that an elevation sufficient to prevent
the deposition of any Pierre shale could have taken place at Golden,
while a few miles to the north and south the true thickness of the
Pierre (7,700 feet according to Eldridge) was deposited.
14. Another weak point in the arch hypothesis is the present
attitude of the strata. These if considered in the plane of their
bedding form a syncline with east-west axis over the crest of the
supposed arch. ‘The limbs on each side are dipping inward as much
as 35.. None of the strata show any evidence of the original anti-
clinal folding to which they must have been subjected in order
to form the “arch.” A simple calculation will show that the arch
requires an anticline with average dips of at least 10° in the older
rocks for ro miles north and south of Golden, while in the crest
of the arch the dips must have risen as high as 30°. Eldridge,
Emmons, and Fenneman are therefore driven to the conclusion
that this arch was flattened out probably in Denver time and
distorted into its present shape. This requires the sudden conver-
sion of a persistently and rapidly rising area into a remarkably
rapidly subsiding one, and requires a bending in the strata similar
to that of a card bent back and forth between the fingers. This
whole reasoning is not only laborious but also illogical.
2 ODNGit Da. 32.
728 VICTOR ZIEGLER
The accumulative weight of the objections summarized above
is so great that the writer unhesitatingly rejects the arch hypothesis
as a possible explanation of the major structural features shown
at Golden.t It is weak in its inherent fundamental assumptions,
does not explain the many and peculiar coincidences in the facts
observed along the line of the supposed unconformities, and is in
several cases directly contradicted by dip and strike observations.
Proposed Explanation of the Structure at Golden: The geo-
logical conditions at Golden, outlined above, are best explained
on the basis of extensive faulting practically parallel to the general
strike of the formations. It is logical to believe that intense mono-
clinal folding of the “‘S” type shown to be characteristic of this
part of the foothills could not be localized along a practically straight
line without a decided tendency to form fractures and: faults
parallel to the general trend of the fold. On the accompanying
geological sketch map of the vicinity of Golden the location of such
faults is indicated by heavy lines, and it will be noted that these
coincide with the lines of two of the unconformities of Eldridge
in his arch hypothesis. The accompanying sections show in detail
the faulting as postulated by the writer. In the case of section
A-A, the double fault and its effect on the strata in the monoclinal
fold is shown by a reconstruction of the latter in its original con-
dition preceding each displacement (Figs. 4-7).
It will be noted that the faults are considered to have steep
westward dips. The author has arrived at this conclusion from a
study of the relationship of the course of the fault line to the topog-
raphy, from a consideration of the character of the monoclinal
fold, and the effect of the fault on the displaced rock formations.
Ordinarily a thrust fault with dips as steep as indicated would be
considered unusual, but we must carry in mind the fact that, in
most cases of thrust faulting, lateral pressure is the dominant cause
—as, for example, in the southern Appalachians. Here, therefore,
thrust faults are as a rule characterized by flat dips. In the case
of a monoclinal fold such as this, however, the maximum pressure
t Richardson, U.S. Geol. Survey, Folio 198, p. 11. It is of interest to note that
both Lee and Richardson appear to doubt the truth of the arch hypothesis. In this
connection see Lee, op. cit., p. 32.
729
FOOTHILLS STRUCTURE IN NORTHERN COLORADO
UapfOH JO YINOs saytur moj ynoge ‘auty Jsem-jJseo Suoye uodes oinjon1yG—'S “OIJ
‘ = Se SS a
Wy SSN SS
WOSIIEOW % SNA ca a NIP-LNIOS ~~
K
S
JOH roroET aN Jive 7 ~ SORE CES SS FwyFNd x
‘
YFIANIO
Uspfoy) Ysnory} sul] JsaM-jsva SuOTe UoTIes sINJoNIyS— PF ‘OL
FOHY dob = FINA 7 < STUHAXOF sea Fic’ f
YIANIG
rok
tf 5
i NY FHI
NIVLNIOS
730 VICTOR ZIEGLER
must have been nearly vertical and must have been manifest at the
edges of the uplift in severely crowding the strata upon each other,
owing to slipping and
ae
Morrisey =~ ~~ : sliding on the steeply
(Pe SN inclined Archean floor.
gong Under these conditions
the most logical planes
of slipping would be the
bedding planes, and the
resulting faults would
be strike faults with
steep dips. The general
dip and strike rela-
tionship shows normal
first Fault East of Tunnel House
Fic: 6.—Detail of first fault shown on
Fig. 4.
easterly dips to be the rule in the
older formations to the west of
the fault plane, while the younger
rocks to the east are characterized
by overturned westerly dips. A
i y : Fic. 7.—Detail of second fault
consideration of the monoclinal (eastern one) shown on Fig. 4.
fold will show that only a steep
westward-dipping thrust fault as drawn by the writer can explain
this relationship. Any eastward-dipping fault with overthrust
from the east would bring younger horizontal rocks to rest upon
FOOTHILLS STRUCTURE IN NORTHERN COLORADO 731
older steeply dipping ones, which is not the case. As will be seen
from the sections, these faults are of the hinge type, with maximum
displacements at Golden. This gradually decreases to zero toward
the north and south.
All facts recorded by Eldridge and all facts observed by the
writer accord perfectly with such an explanation. We should
expect fault lines of this nature to be characterized by the crumpling
and overturning of strata affected. ‘The more sharply a formation
is truncated the more evident the effects on its bedding planes should
be. Maximum displacement and maximum disturbance should
logically go hand in hand. The course of the fault plane and its
position would determine whether a formation disappears from the
bottom up or from the top down. As a general rule, with dips and
strikes as observed, the formation on the west side of this fault
plane should disappear from the top downward, while the formation
on the east side should disappear from the bottom up. This is
actually the case (Fig. 8).
Upon cursory examination the decided westward curve of the
outcrops is somewhat surprising and seems to suggest that the
eastern block represents the upthrow side. This is, however, not
true, and the inward curve is the combined result of the gradual
steepening and eventual overturning of the monoclinal fold.as we
approach Golden, and the displacement along the westward-
dipping fault surface.
The actual inward sweep of the strata resulting from the writer’s
interpretation of the structure can readily be approximated as
follows: The total thickness of strata actually cut out at Golden is
about 10,000 feet. Therefore, the lowest bedding plane on the
Fox Hills remaining at Golden must have been located at least
10,000 feet higher than its present position on the monocline before
faulting. To this must be added the difference in elevation between
the top of the Archean and the Fox Hills today (at least 1,200 feet)
and a certain amount to allow for folding, hence 12,000 feet may
be considered a safe estimate as to the minimum amount of throw
necessary to bring the base of the Fox Hills from its original loca-
tion in the monocline to its present position. A vertical drop of
12,000 feet on a fault plane dipping 55° westward will result in a
VICTOR ZIEGLER
fhe
westward travel of 8,000 feet on the horizontal plane. The result
of the displacement alone will be a decided westward heave in any
Surtace
QIAGRANM
Jo sllustrate disappearance
of beds along “he saul? pare.
With ricreasing arsplacermen?
beds a7 wes? side asageear
trom the top dowrr; beds on
the east side, Sram Loffarm yo.
Fic. 8.—Diagram illustrating the disappearance of beds along the fault plane.
Older beds (on left) disappear from top down; younger beds (on right) disappear
from bottom up.
outcrop, and, since the amount of westward travel is dependent
upon the amount of throw, each outcrop should show a progressively
larger travel toward the west as we approach Golden.
FOOTHILLS STRUCTURE IN NORTHERN COLORADO — 733
This westward travel of 8,000 feet, due to the fault, will be
augmented considerably by the change in the character of the
‘monoclinal fold. As has been shown above, the monocline north
and south of Golden is normal, gradually steeping and eventually
overturning as it approaches Golden. The natural dip of the
formations in the normal parts would, therefore, carry their out-
crops far to the east of the steeper and overturned parts of the
monocline, which, added to the effect of the faulting, is undoubtedly
sufficient to account for the present situation of the various forma-
tions.
In connection with the writer’s interpretation the following
~ statement from Marvine' referring to the condition at Golden is of
interest: ‘‘Some of the facts at hand indicate that a peculiar
fault, depending on the nature of the sharp fold, and possibly
connected with the lava near by, may have caused the present
appearance.”’
The true nature of the fault was not realized by Marvine, for
he also states “this may be caused .... by a fault which has
pushed the higher portion of the series westward over the upturned
edges of the lower portion, thus concealing much of the latter.”
Such a displacement is, however, incompatible with the data at
hand and was hence rejected by later workers.
It is also of interest to note that Lee and Richardson’ appear
to doubt the existence of local unconformities in the foothill region.
Dr. Patton also states in personal conversation that many of
the phenomena observed by him in the foothills appear to be
incompatible with the arch hypothesis.
The Boulder area——The same structural peculiarities are
shown at Boulder as at Golden, but not developed to the same
remarkable degree and differing in some minor detail. Eldridge,’
and subsequently Fenneman,’ studied this area in detail and
developed an explanation in every respect similar to that advanced
for the conditions at Golden (Fig. 9). Elsewhere’ the writer
t Hayden Survey, VII (1873), 137, 138.
2 See Lee, op. cit., p. 32; Richardson, U.S. Geol. Survey, Folio 108, p. 11.
3 U.S. Geol. Survey, Monographs, XXVII, 105-14.
4U.S. Geol. Survey Bulletin 265, pp. 54-66.
5 Colo. School of Mines Quarterly, April, 1917.
734 VICTOR ZIEGLER
has discussed the Boulder area in detail and has shown that in
this district also it fails to account for the local structural features
and is fundamentally as unsound here as it is at Golden.
The accompanying sections show the writer’s interpretation
of the structure at Boulder. Here, as at Golden, steeply dipping
strike faults have taken place parallel to the general trend of the
foothills monocline, which are responsible for the disappearance of
some formations and the notable decreases in thickness exhibited
by others. The writer will not attempt to discuss these faults in
Fic. 9.—Sketch map of vicinity of Boulder (after Eldridge). Heavy lines indi-
cate faults. For symbols, see Fig. 2.
detail. They resemble in every respect the faults at Golden and
affect the strata in a similar way, but to a lesser extent. They
are in absolute harmony with the manner of disappearance and
thinning of the various formations, with observed dip and strike
relationships, and do not postulate the extreme local subsidences
and elevations required of the older explanation (Fig. 10).
Minor structural features —Much minor faulting and folding,
some of it of eccentric type, characterizes the great monoclinal
uplift. Elsewheret the writer has discussed this in some detail,
especially for the purpose of bringing out the relation existing
Op. cit.
FOOTHILLS
STR
UCTURE IN NORTHERN COLOR
ADO
735
5
Se
S1 S
| | y
Aaa
ah
wih nl
eS
| i lil HI qk S
Ha 3
r i ; i n°
SEMEL |
SLs 3
SMES
Ny & é
aaa) pos
a) E
ATH A | “Eb
Tae | E
oo i = :
Oy, Pane S
We CoS 3
Seo = : a
f ee :
S aes
Vv A0y z
<=
SS
4 *
oy S
(a OR
~
o~ ford 4n~7
4 OEE Pe yi SNe
Fines 7 a5 = 7 N N= a
SA ALAN Ge ESE nil OR
NG ioe Ne SST S
sw
736 VICTOR ZIEGLER
between these minor structures and the huge thrust faults here
described (Figs. 11-15).
Minor Folds: Two distinct types of minor folds can be recog-
nized, folds en échelon and drag folds. ‘The former pass practically
invariably on their west flank into eastward-dipping faults with
strikes nearly parallel to the axis of the fold. They represent
v
v
6
oa
b
v
e
v
v
v
bh
Fic. 11.—Geological sketch map of area on Estes Park road west of Loveland,
showing characteristic fold en échelon and accompanying thrust fault. Section lines
are indicated.
minor wrinkling, with overthrusts from the east subordinate in
amount to the main monoclinal uplift and faulting. Other folds,
noted especially in the shale series, represent adjustments by in-
competent layers to stresses incident to the formation of the
uplift.
Minor Faults: Both strike and dip faults are numerous, espe-
cially in the structurally disturbed areas at Golden and Boulder.
They are all of slight displacement. They do not antedate the
\FOOTHILLS STRUCTURE IN NORTHERN COLORADO 737
uplift of the monoclinal fold, as stated by Eldridge, Fenneman, etc.,
but represent minor fracturing and slipping attendant upon, and
consequent to, the formation of the main monoclina! uplift and
its master-faults.
x
NOD
N
23
fran
a
eee
meee
Sa
[cana =a)
feetenesa|
eer rit
aaa
ESD.
eae ZaN
eaeaTaEN
samen
PREERSA TERN
B77
HAV oRaeSEA
RSE,
eee
Fic. 12.—Geological sketch map of southwest corner of Loveland quadrangle,
showing folds ex échelon. For symbols, see Fig. 2.
SUMMARY
The master-monocline.—It has been shown by a reconstruction
of the original foothills fold from observed dip and strike relation-
ships that this represents an ideal monoclinal fold in strata origi-
nally continuous over the entire area of the Front Range, which
gradually steepens and eventually overturns as it approaches
Boulder and Golden. Contrary to former interpretations and
general belief, the maximum overturn is above the earth’s surface.
The main faults.—It has definitely been proved that the struc-
tural irregularities at Golden (and Boulder) cannot be explained
738 D)) AVICTOR ZIEGLER
=
S
x ul
8 ar
S
rs iM
meigy
We
Fic. 13.—Section along line indicated on Fig. 12
1
Ves
/
aN
ZR IAN
SPAXKS
Sea Level
Fic. 14.—Huge fold en échelon and fault west of Loveland, east-west section, through Mt. Milner
FOOTHILLS STRUCTURE IN NORTHERN COLORADO 739
as the result of repeated local arching and contemporaneous
erosion, as attempted by former workers in this field. An alter-
native interpretation has been advanced by the writer which
(SIAN SRG
ys liz
IAN
\
~~
oN
cS
(
Se
(
r
\N
NilENS
eN
a
ae
Fic. 15.—Fold ex échelon of the Twin Peaks at Boulder. East-west section
Lary.
On ®@ Go/den \ \
P Ry Main Thrust Faclfp --F
J Fras. Minor Dip fars/ts-Ff. f.--
Qagramn
76 (llustrate cause
Atvota/l Aip ~
Se
a
Fic. 16.—Diagram to illustrate the displacement of the dip faults and their rela-
tion to the large thrust faults at Golden. Small diagram shows distortion of inter-
fault block due to pressure causing the main fault. The small arrows indicate the
direction of displacement of interfault blocks.
740 VICTOR ZIEGLER
postulates the presence of extensive strike faults with steep westerly
dips and overthrust from the west. This interpretation is shown
to be free from the inherent weaknesses of the older so-called
arch hypothesis, and to be in perfect accord with all observed
geological features.
Conclusion.—The formations represented in this area of the
foothills range in age from Permo-Carboniferous to Paleocene.
The Cretaceous formations certainly, the older probably, extended
formerly over the entire area of the Front Range. At the close of
Arapahoe time and during early Denver time the area to the west
of the foothills rose, and the sedimentaries were folded into a normal
monocline with average eastward inclination of about 45°. Locally
(as at Golden and Boulder) excessive overturning occurred, accom-
panied by fracturing, and resulting in extensive overthrusts from
the west along steeply dipping fault planes. Minor strike faulting,
and dip faulting with hingelike displacements, as well as drag fold-
ing, accompanied and followed the major uplift and faults. The
natural buckling and wrinkling of strata, where the pressure on
the monoclinal fold was not intense enough to cause huge strike
faults, resulted in the formation of folds ex échelon, with their
attendant overthrust faults from the east. The exact time of the
uplift cannot be closely determined, except that the basal beds of
the Denver are involved in the folding and that the composition
of the upper beds of this formation shows that erosion had entirely
cut through the sedimentary series over the site of the Front Range
at the time of their deposition.
The sincere thanks of the writer must be expressed to Dr. H. B.
Patton for advice and helpful suggestions in the preparation of this
paper.
ON THE GEOLOGY OF THE ALKALI ROCKS IN
THE TRANSVAAL
H. A. BROUWER
Delft, Holland
CONTENTS
INTRODUCTION
IGNEOUS COMPLEX OF THE BUSHVELD
Place of the Main Transvaal Laccolith
Bottom of the Laccolith
Roof of the Laccolith
Accompanying Dikes and Intrusive Sheets
Tectonic Changes Connected with the Intrusion
Contact Metamorphism
Rock Types of the Bushveld Complex
Pneumatolysis
THE OCCURRENCES OF ALKALI ROCKS
a) The Pilandsberg
General Character of Rocks
The Rocks of the Country around the Pilandsberg
Dike Rocks
Pegmatites
Mechanism of Intrusion of the Pilandsberg Complex
Age of the Pilandsberg Complex
b) Other Occurrences of Nepheline Syenites and Allied Rocks
The Intrusion.on Leeuwfontein (320)
Nepheline Syenite Region to the West of Lydenburg
c) Origin and Age of the Nepheline Syenites and Allied Rocks
INTRODUCTION
In June 1910 I studied the geology of the occurrences of nephe-
line syenites in the Transvaal, and the results were published in the
same year in a paper, “‘Oorsprong en Samenstelling der Trans-
vaalsche Nepheliensyenieten,” which contains a contribution to the
geology and a petrographical description of the nepheline syenites
and various allied rocks. In the following pages some geological
questions will be treated more in detail and the main results of
recent work of the Geological Survey of South Africa are added.
741
742 H. A. BROUWER
I am much indebted to Professor G. A. F. Molengraaff, of Delit,
who placed his collection of rocks at my disposal, while the text
has been much improved by his valuable suggestions; and also to
Professor A. Lacroix, of Paris, the rock material being studied for
the greater part in his laboratory.
For assistance during my stay in South Africa, my best thanks
are due to the staff of the Geological Survey, particularly to Mr.
A. L. Hall, who furnished some rock specimens from Rietfontein
[451] and Spitskop [463].
As a guide to the stratigraphy and the geological dates men-
tioned below, a table of the Transvaal formations is entered (see
Table I).
IGNEOUS COMPLEX OF THE BUSHVELD
The nepheline syenites of the Transvaal make part of a complex
of igneous rocks, which is intrusive as a laccolith or laccolithic
sheet between the upper strata of the Transvaal system and
the strata of the unconformably overlying Waterberg system.
The laccolithic character of this complex was recognized by
Molengraaff,? who grouped them under the name “‘plutonic series
of the Bushveld’’;3 it includes igneous rocks, which have a high
soda content as a common character. The name used by other
authors, “igneous complex of the Bushveld” or ‘Bushveld
igneous complex,”’ has the same signification and can be considered
as the official one, because the Geological Survey of the Transvaal has
adopted it.
PLACE OF THE MAIN TRANSVAAL LACCOLITH
The western boundary of the part of the complex in the Central
Transvaal, which has been uncovered by erosion, is found nearly
15 km. to the west of the Marico River; the eastern one,
nearly 25 km. to the west of Lydenburg; the medium breadth is
nearly too km.
But probably it covers a much larger area; at least it appears
again between the Magalakwin River and the sources of the
t Throughout this paper numbers in brackets refer to official designations for
farms in South Africa.
2G. A. F. Molengraaff, ‘‘Géologie de la République Siid-Africaine du Transvaal,”
Bull. de la Soc. Géol. de France, Série 4, T. I, 1901, p. 13.
3G. A. F. Molengraaff, Geology of the Transvaal (Johannesburg, 1904), p. 42.
ALKALI ROCKS IN THE TRANSVAAL 743
Matlabas and to the north of the Palala Plateau, where it extends
to the north to near the Limpopo, nearly 30 km. to the south of the
northern boundary of the Transvaal.'
BOTTOM OF THE LACCOLITH
The bottom of the laccolith is formed everywhere by the upper
strata of the Pretoria series, generally consisting of Magaliesberg
quartzite.
Therefore the rocks of the laccolith in the Central Transvaal
are for the greater part surrounded by the upper Pretoria strata;
in the strata of the Transvaal system, which dip everywhere
toward the central part of the complex, the dip of the strata de-
creases when the distance from the complex increases. Because
the Magaliesberg quartzites of the Pretoria series have specially
resisted erosion we now see them as a ridge surrounding the complex.
We see that the bottom of the laccolith is determined, but the
the place of the roof of the laccolith is uncertain.
ROOF OF THE LACCOLITH
The part of the laccolith, which has not been uncovered by
denudation, is covered by the strata of the Waterberg system and
partly by the younger strata of the Karroo system. Between the
basal conglomerate of the sandstone series of the Waterberg
system and the underlying rocks of the Bushveld complex we some-
times find a series of felsitic rocks, which other authors have con-
sidered as being directly connected with the deep-seated rocks of
the laccolith, but which the Geological Survey of the Transvaal
regards as a lower division of the Waterberg system.
The Waterberg system then includes:
Upper Sandstones, grits, and
Division conglomerates
Waterberg system 4 Lower Felsites and allied volcanic
Division rocks with interbedded
Prin | shales
1G. A. F. Molengraaff, ‘‘Geologische Aufnahme der Siid-Afrikanischen Repub-
lik,” Jahresbericht tiber das Jahr 1898, Pretoria, 1900; G. G. Holmes, ‘‘Some Notes
on the Geology of the Northern Transvaal,” Trans. Geol. Soc. South Africa, VII (1904),
51-56.
744 H. A. BROUWER
The roof of the laccolith would, then, not be formed by the
sandstone series of the Waterberg system but by the volcanic
TABLE I
TABLE OF THE GEOLOGICAL FORMATIONS OF THE TRANSVAAL!
Main Igneous Intrusions Sedimentary Rocks Age
Superficial Deposits
Unconformity
Upper (Bushveld amygdaloid ?
Karroo {Bushveld or Buiskop ?
é sandstone
Middle (Highveld or
ae Karroo [Beso series with
coal measures Permo-
Lower ({Ecca shale series Carboniferous
Karro 4Dwyka conglomerate
or tillite
Bushveld igneous com- Waterberg system
plex intruded as Unconformity
an unconformational],, Pretoria series ;
laccolithic sheet be- Pere: Dolomite series Probe aly faa
tween the strata of| ** Black Reef series
the Pretoria series
and those of the
Waterberg system
Unconformity
Ventersdorp or Vaal River system ?
Old grey granite-—The Unconformity Nae
exact age of the in-|__. Upper division, containing
trusion or the intru-|W2twatersrand the famous _ auriferous ?
sions of this granite] System Main Reef series
is not yet known. Lower division ?
The intrusion, how-
ever, took place cer- Unconformity
tainly after the de- Swaziland system fy
position of the rocks
of the Swaziland sys-
tem and before the
deposition of the
rocks of the Trans-
vaal system
series, in which sometimes red granites can be seen clearly intrusive.
But the question of the mutual relation between the rocks of the
Cf. G. A. F. Molengraaff, ‘‘The Deposits of Iron Ore in the Transvaal,” in The
Tron Resources of the World (Stockholm, 1910), p. 1060.
745
ALKALI ROCKS IN THE TRANSVAAL
St] uroquojjary Qy <[911] Tepsueuypem ‘M ‘[9zt] yroodsuvagq ‘yy {[611] uteyuoysodsoT ‘rT ‘[96€] [eerymnaay] ‘rT ‘[oz€] uraywoyansa'T “J :Br1aqspuettg 94} Jo xajdu04 y,
«SYoor pore pur soyuacs ourpaydeu Jo saouarmM990 oy} YIM [evasueIY, [eIJUAD oY} Jo deur yoyoys [eotsojoor)— TI “oy
W9}SAS [VVASUVIT,
w1a}sAs [VVA ae
-SUPIT, 94} UY? soyxIp u19}sAs
JOp[o Stlol}eUIIOF ‘oqo ‘soz sojruaAs pur Solas Sorias ayuaAs YIM jevasuriy, 94} ueygy
pue soyueizs pjQ -uaxo1Ad ‘so}ION SOURIS pry sa}iuaAs sutjayda nN Pay PLT d}IULOJOC, ‘salias VUOJIIG JasunoA SuUOI}eULIO
| Baie ee rear
i ie
quays
ea
44 4h phils
+, 4,0 sp, — 1
+ RES —1|
an
yl ly
my
* ie Oe 4
"Stel qe
{ eee
ays oe
ot
rad
746 H. A. BROUWER
Bushveld complex, the volcanic series, and the sandstone series, of
which the contradictory facts already were enumerated by Molen-
graaff,’ has not been solved in this way.
As far as concerns this mutual relation, the following facts can
be considered as certain:
t. The deep-seated rocks are intrusive in the volcanic series.
2. Dikes which are genetically connected with the deep-seated
rocks (felsophyres with the red granites, tinguaites with the
nepheline syenites) cut through the sandstone series.
3. The volcanic series show the characteristics of effusive rocks
and include sediments (shales and sandstones) in the higher
horizons.
4. Sometimes more, sometimes less, clearly the sandstone series
rest unconformably upon the volcanic series, the transition being
characterized by the existence of conglomerates.
5. In the basal conglomerate of the sandstone series felsitic
pebbles sometimes occur.
6. Fragments of a conglomerate, which very closely resembles
the basal one, are found in the phonolites of the Pienaars River
valley.
When we try to make these facts agree with each other, we meet
with great difficulties. It seems to be certain that an effusive
period has preceded the main intrusion and that both are connected
genetically.
But when we admit, for instance, that the sandstone series are
older than the effusive and intrusive series, then the occurrence
of felsitic pebbles in the basal conglomerate and the more or less
distinct unconformity with regard to the felsites, and also the effu-
sive character of the latter ones and their alternation with sedi-
ments, are inexplicable. When we admit that the sandstone series
are younger than the effusive and intrusive period, then the inter-
secting dikes, which are the equivalents of the deep-seated rocks,
and also the intrusions of red granite in the sandstone series are
unexplained.
We could explain the facts in a rather satisfactory way if we
admitted effusion and several intrusions from.a deeper-seated
1G. A. F. Molengraafi, Geology of the Transvaal (Johannesburg, 1904), p. 59.
ALKALI ROCKS IN THE TRANSVAAL 747
mother-magma and considered the Waterberg sandstone as younger
than the effusive, but older than the main intrusive, period; then,
however, there is no genetic connection between effusion and
intrusion.
With H. Kynaston and E. T. Mellor,’ we can admit a prolonged
activity of the Bushveld magma, while the sandstone series were
deposited between the main intrusive and effusive period and the
later intrusions.
That the nepheline syenites are younger than both the norites
and the granites will be proved in the following pages. Only
detailed geological and petrographical investigations can clear the
true succession from the remnants left after the advanced denuda-
tion.
In any case, the following reasons make it undesirable to unite
the volcanic series and the sandstone in the Waterberg system.
Over large surfaces at the base of the sandstone series we find
developed a basal conglomerate, which rests unconformably, not
only upon the felsites, but also upon the red granites; and in the
Zoutpansbergen we also find them resting upon much older forma-
tions, such as the old granites. ‘Thus it forms a geological horizon
over the whole of Central Transvaal and marks an extensive
unconformity.
As a rule the pebbles of felsites are rare in this conglomerate,
although the relative quantity may increase locally and the uncon-
formity be locally less well developed. In the Waterberg district?
the conglomerate is principally composed of pebbles of jasper,
quartzite with magnetite, white quartzite, schistose quartzite with
muscovite, quartz, chert, and felsophyre, which, except the fel-
sophyre, belong to the Swaziland system (Barberton series). Con-
sequently this unconformity must be maintained as a horizon of
separation, and the volcanic series must be separated from the
sandstone series of the Waterberg system and can either be included
in the igneous complex of the Bushveld or else considered as a
tH. Kynaston and E. T. Mellor, The Geology of the Waterberg Tin Fields. Memoir
No. 4, Geological Survey of the Transvaal, Pretoria, 1909.
2G. A. F. Molengraaff, ‘‘Geologische Aufnahme der Siid-Afrikanischen Repub-
lik,” Jahresbericht tiber das Jahr 1898, Pretoria, 1900.
748 H. A. BROUWER
separate volcanic series, which is younger than the Transvaal
system and older than the Waterberg system and the main lacco-
lithic intrusion.
ACCOMPANYING DIKES AND INTRUSIVE SHEETS
Except the basic rocks, which alternate with the shales and —
quartzites of the Pretoria series and which are perhaps genetically
connected with the intrusion, numerous syenite dikes cut through
the Pretoria series and the Dolomite series, and intrusive sheets
of red and grey syenite are found in the dolomites. The well-
known dike of porphyritic nepheline syenite of the station Wonder-
fontein in the Potchefstroom district can be followed over Breedts
Nek in the Magaliesbergen as far as the nepheline syenites of the
Pilandsberg.
At the contact of the intrusive sheets, which have a thickness of
three to forty meters, the syenite is finer-grained to microcrystal-
line, and the dark dolomite has been changed into white marble.
TECTONIC CHANGES CONNECTED WITH THE INTRUSION
The study of the Transvaal system in the neighborhood of the
laccolith proves that there are numerous dislocations directly
connected with the intrusion.’
The strata sank under the weight of the intrusive mass; this
explains the increasing of the dip, when the distance from the
complex decreases, and also explains why the complex is sur-
rounded by a ridge of harder sediments, which dip toward the
central part. In the neighborhood of Pretoria and from there to
the west, as far as Rustenburg, we see the ridge of the Magaliesberg
quartzites uninterrupted, the strata dipping toward the intrusive.
complex.
To the east of Pretoria is a series of step faults, which can be
followed easily in parallel ridges, which consist of quartzites of the
Pretoria series.
We see the dislocations in a remarkable manner where the
periphery of the complex forms a re-entering angle as, for instance,
at Franspoort east of Pretoria. The ridge of the Magaliesberg
‘This question was discussed in detail by Molengraaff; cf. Geology of the Trans-
vaal (Johannesburg, 1904), p. 50.
ALKALI ROCKS IN THE TRANSVAAL 749
quartzites and the accompanying Daspoort and Timeball quartz-
ites here suddenly bend to the southeast. The dip of the strata
continues toward the red granites, but the outer ridge of the Maga-
liesberg quartzites has been fractured and extended in length; the
ridge is broken by “‘poorten.” The inner ridges (Daspoort and
Timeball Hill quartzites) were strongly pressed in a direction
slightly oblique to the strike of the strata. All this is clearly shown
by the grouping of the quartzite hills in the neighborhood of Pre-
toria. That the intrusion and dislocations are directly connected
is also evident from the study of the zones of contact in the sur-
rounding sediments in disturbed and undisturbed regions.
CONTACT METAMORPHISM
The quartzites, clay-slates, and “greywackes”’ of the Trans-
vaal system are strongly metamorphosed by the intrusion of the
laccolith. The contact phenomena in connection with the laccolith
were first mentioned by Molengraaff and later studied in detail
by Hall.t The quartzites of the Magaliesberg Range are recrys-
tallized and consist of more or less hexagonal quartz crystals, which
sometimes attain a diameter of more than one centimeter; in the
clay slates cordierite, andalusite, and biotite appear. The meta-
morphism decreases when the distance to the laccolith increases,
but even the rocks of the dolomite series are metamorphosed.
Where the Transvaal system is much disturbed and has undoubt-
edly been exposed to high pressure, the metamorphosed rocks show
a different character in their structure, as well as in their mineralogi-
cal composition. These rocks are connected by transitions with
the pure contact-rocks of the undisturbed regions.
Muscovite, glaucophane, and zoisite, which are characteristic
for the dynamometamorphic crystalline schists, and in small
quantities the contact minerals, cordierite, andalusite, and tour-
maline, occur in the metamorphosed rocks. Hall decides upon the
contemporaneous action of contact and dynamometamorphism in
the disturbed regions, from which it is once more evident that the
intrusion was directly connected with the dislocations.
tA. L. Hall, ‘Uber die Kontaktmetamorphose an dem Transvaalsystem im
éstlichen und zentralen Transvaal,” Tschermaks Min. u. Petr. Mitt., Bd. XXVIII,
Heft 1-2 (1909), pp. 115-52.
750 H. A. BROUWER
ROCK TYPES OF THE BUSHVELD COMPLEX
1. Red granites—The acid rocks of the laccolith are amphibole-
biotite granites, which are very poor in dark constituents. Prin-
cipally they show typical granophyric structure. They differ
petrographically from the old granites, in essential features enum-
erated by Molengraaff.t In the red granites muscovite is entirely
wanting.
2. Norites, gabbros, and pyroxenites (with segregations of tron
ore).—Nearly everywhere at the periphery of the red granites we find
a zone of basic and ultra-basic rocks. The basic rocks are found
near the western and southern part of the Pilandsberg; they
accompany the Magaliesberg quartzites in a south-southeasterly
direction to the environs of Rustenburg, where they bend to
the east in the direction of Pretoria. The Zwartkoppies and
Pyramids have been given their respective names from the color
and the form of the small hills, which are composed of these rocks.
Finally, they are found from the environs of Belfast to those of
Piet Potgietersrust; still farther to the north they are in contact
with the old granites. Iron ore has been segregated from these
basic rocks at several places; lenticular masses of magnetite are
developed nearly everywhere around the Bushveld. At some places
these masses are thicker than one hundred meters. The iron ore
is magnetite, sometimes with chromite. In the norites the per-
centage of magnetite goes on increasing, as one approaches the
pure magnetite. Ultra-basic pyroxenites and peridotites fill the
shallow basin to the West of the Pilandsberg, bounded on the south
by the Schurveberg and the Zeerust Hills, on the west by the Marico
Hills, and on the north by the Dwarsberg.
3. Nepheline syenites and syenites.—These rocks, of which the
mode of occurrence and the composition are more fully described
in the following pages, are uncovered at several places, often near
the boundary of norites and granites.
PNEUMATOLYSIS
As a result of the cooling down and contraction of the intrusive
complex and the pressure upon the surrounding strata, numerous
1G. A. F. Molengraaff, Geology of the Transvaal (Johannesburg, 1904), p. 44.
ALKALI ROCKS IN THE TRANSVAAL 751
fissures were formed, in which different minerals crystallized from
the circulating emanations of the magma. The numerous occur-
rences of tin ore are genetically connected with the red granites
of the Bushveld. The copper and silver ores of the Albert mine and
the cobalt ores of Balmoral are found in red granite. Finally, we
find numerous ore deposits in the Pretoria series, the origin of which
is probably directly connected with the intrusion of the Bushveld
complex."
THE OCCURRENCES OF ALKALI ROCKS
Common to the occurrences of alkali rocks is the abundance of
nepheline syenites. They are exposed at the following places:
1. In the Pilandsberg (Rustenburg district), where they are
accompanied by syenites, effusive rocks, and dike rocks.
2. At the boundary of the farms Leeuwfontein [320] and
Zeekoegat [287], northeast of Pretoria, where they are accompanied
by syenites, leeuwfonteinites, effusive rocks, and dike rocks.
3. On the farms Rietfontein [451] and Spitskop [463], west of
Lijdenburg.
4. On the farm Franspoort [426], south of Leeuwfontein [320]
(Pretoria district).
5. On the farm Walmansdal [116], northwest of Leeuwfontein
[320].
6. On the farm Leeuwkraal [396], still farther to the northwest.
7. On the farm Losperfontein [119] (Rustenburg district).
8. Numerous dikes can be followed from the Pilandsberg in a
northwesterly and a southerly to southeasterly direction.
tG, A. F. Molengraaff, Geology of the Transvaal (1904), p. 52; A. L. Hall, ‘‘Geo-
logical Notes on the Bushveld Tin Fields, etc.,”’ Trans. Geol. Soc. South Africa, VII
(1905), 47-55; F. H. Hatch and G. S. Corstorphine, Geology of South Africa (London,
1909), p. 216; E. T. Mellor, ‘‘ Field Relations of the Transvaal Cobalt Lodes,” Trans.
Geol. Soc. South Africa, X (1907), 36; H. Kynaston, ‘‘Anniversary Address of the
President of the Geological Society of South Africa for 1908,” Trans. Geol. Soc. South
Africa, XII (1909); H. Recknagel, ‘‘On Some Mineral Deposits in the Rooiberg
District,” Trans. Geol. Soc. South Africa, XI (1908), 83; ‘‘On the Origin of the South-
African Tin Deposits,” Trans. Geol. Soc. South Africa, XII (1909), 168; H. Merensky,
‘““The Rocks Belonging to the Area of the Bushveld Granite Complex in Which Tin
May Be Expected, etc.,”’ Trans. Geol. Soc. South Africa, XI (1908), 25; H. Kynaston
and E. T. Mellor, ‘‘The Geology of the Waterberg Tin Fields,” Memoir No. 4, Geologi-
cal Survey of the Transvaal, Pretoria, 1909.
752 H. A. BROUWER
THE PILANDSBERG
Where the ridge of the Magaliesberg quartzites—the direction
of which was southeast and northwest, to the north of Rustenburg—
bends again to the west, a mountain group arises from the rolling
| Coole: geal Midhrmiot
of
Bey
i hash
kane spruit tain saree sal
38 Kru Gey
Se wane
Kilometers:
Zale Veeco eee cease of Me pe gpnl
Sek Pica +n He Marat “Borord for Guts
Fic. 2.—Geological sketch map of the Pilandsberg
and undulating country of norites and granites. It has an almost
circular outline with a diameter of 15 miles from north to south
and 18 miles from east to west. Long ridges and isolated mountains
are separated by broad valleys, in which small rivers, dried up in
winter, have cut deep valleys with vertical walls.
ALKALI ROCKS IN THE TRANSVAAL 753
The first explorer who mentioned the composition of the Pilands-
berg and the surrounding region was Adolf Hiibner, in his descrip-
tion of a voyage from Potchefstroom to Inyati.t Evidently he
had observed the peculiar character of these rocks, because he
writes: ‘Am meisten Beachtung verdient wohl das unten zu
beschreibende Gestein der Pilandsberge, welches entschieden als
ein basisches plutonisches Gestein zu den Griinsteinen gerechnet
werden muss.”’ He crossed the Magaliesberg Range along the
Hex River at Olifants Nek and traveled from there to Morgenzon
[427] (the same direction which the road from Rustenburg to the
Pilandsberg still follows); as the rock of the plain of Rustenburg
he mentions a typical medium-grained greenstone, which near the
contact with the rocks of the Magaliesberg Range crops out in
thick sheets. Then he crossed the Elands River and reached the
Pilandsberg, which, he says, consists of mountains of greenstones
from about 400 to 600 feet high. The meaning of his sentence,
“Die Gesammtheit der Quarze bildet ein wahres Massengebirge,”’
is not very clear. He mentions that a part of the mountains con-
sists of a rock which seems to be a “‘hornblende porphyry,” but
which shows a syenitic character on closer examination. Because
it consists of two minerals, a red “‘felsite’’ and a black amphibole,
Hiibner says that it is not a normal porphyry, though the red feld-
spar (orthoclase) predominates. The amphibole does not form
crystals, but shows rather regular forms. As an interesting feature
of this perhaps quite new rock, he mentions the numerous inclu-
sions of clay slate and granite, which do not show any contact
metamorphism. According to Hiibner, this rock would cover a
large area in the Pilandsberg. At several places, e.g., behind the
negro town on Saulspoort [369] in the northeastern part of the
mountains, granites and eruptive breccias, which contain fragments
of porphyrite and granite, occur. Because Hiibner mentions
that he visited the missionary station on Saulspoort, he perhaps
speaks about these rocks, which occur in the neighborhood of
syenites with red feldspars.
t Adolf Hiibner, ‘‘Geognostische Reisen in Siid-Africa,”’ Peterm. Geogr. Miit.,
XVIII (1872), 424, 420.
2 These rocks probably are the norites with schistose structure in the margin of
the igneous complex of the Bushveld.
754 H. A. BROUWER
Nearly at the same time as Hiibner, Carl Mauch’ made geog-
nostical studies in the central part of South Africa. In a chapter,
“Mein erstes Jahr in der Transvaal Republik,” he describes his
“Trrfahrten in den Pilaans-Bergen.”? Coming from Rustenburg,
he passed a conical hill and entered the central part of the mountain
complex and found “quartz porphyries”’ with a violet-brown
groundmass in the bed of a rivulet. After nearly perishing with
hunger and thirst, he was received hospitably in the house of the
missionary, who still lives in the native stadt on Saulspoort. From
the western part of the mountains he mentions pieces of copper
ore, magnetite, fluorine, and pebbles of gneiss. Most probably
his quartz porphyries are the rocks without quartz, with por-
phyritic feldspars in a reddish, dense or fine-grained groundmass,
which are the effusive equivalents of the syenites and nepheline
syenites of this region. His gneiss probably is the schistose lujaur-
ite which covers large areas in the western and southern part of
the mountains.
In his sketch of the South-African Republic, G. A. F. Molen-
graaff? gives a review of the knowledge about the rocks of the
Bushveld, and here the results of Mauch and Hiibner are mentioned.
In 1898 J. A. L. Henderson described a syenitic rock from the
Pilandsberg as pilandite. The true character of the rocks of the
Pilandsberg was recognized by Molengraaff* in 1904. In a short
geological description of the Pilandsberg and a part of the Rusten-
burg district he mentions that different varieties of foyaites are of
widespread occurrence and the schistose varieties, which are very
rich in aegerine, are compared with the lujaurites of Greenland.
The rock specimens, which Molengraafi collected, were studied
by me, and the results of this study were published in the petro-
t Carl Mauch’s “Reisen im Innern von Siid-Afrika, 1865-1872,” Erginzungs-Bd.
VIII (1873-74), Peterm. Geogr. Mitt. No. 37.
2G. A. F. Molengraaff, “Schets van de Bodemgesteldheid van de Zuid-
Afrikaansche Republiek,” Tijdschr. Kon. Aardr. Gen. (Leiden, 1890), p. 604.
3 J. A. L. Henderson, Ox Certain Transvaal Norites, Gabbros, Pyroxenites and Other
South-African Rocks. London, 1898. Dulau and Co.
4G. A. F. Molengraaff, ‘‘Preliminary Note on the Geology of the Pilandsberg
and a Portion of the Rustenburg District,” Trans. Geol. Soc. South Africa, VIII (1905),
108,
ALKALI ROCKS IN THE TRANSVAAL 755
graphical part of my first paper. In the winter of 1910 I visited
the occurrences of nepheline syenite in the Transvaal, especially
that of the Pilandsberg (during the last two weeks of June).
After this visit I was able to give a brief description of the
general geology of the neighborhood.' I showed that these moun-
tains consist of red syenites, nepheline syenites, and their por-
phyritic and dense equivalents, and, with a sketch map, that the
main rock types are disposed in concentric circles. As far as con-
cerns the red syenites, I was able to show that they often are covered
by the effusive rocks, while they probably were older than the
nepheline syenites of the complex. There was clear evidence that
the rocks of the Pilandsberg intrusion are younger than the granites
and norites of the igneous complex of the Bushveld; dikes, which
are genetically connected with the intrusion, cut through the
granites and norites. The intrusion of nepheline syenites on
Leeuwfontein [320] was shown to be certainly younger than the
Waterberg system, because a dike of tinguaite, which has the same
chemical composition as the normal foyaites of Leeuwfontein,
cuts through Waterberg sandstones and conglomerates on Paarde-
fontein [338].
The mapping of the Pilandsberg and surrounding area was
carried out in the next year by Dr. Humphrey, in company with
Dr. P. Wagner, in connection with the work of the Geological
Survey of South Africa.2,_ Humphrey divides the whole of the rocks
into two main groups: the nepheline syenites and phonolites; and
the alkali syenites and trachytes. Each of these groups contains
a plutonic and an effusive representative and ‘‘the reason for this
classification is that the rocks forming the Pilandsberg are the
denuded remnants of what was once a stupendous volcano, com-
parable in size with the greatest of the present-day active
volcanoes.”
General character of rocks.—The foyaites and other allied rocks
in Professor Molengraaff’s collection have been described by me
1H. A. Brouwer, Oorsprong en samenstelling der Transvaalsche Nepheliensyenieten
(1910), pp. 12-29.
2W. A. Humphrey, ‘‘The Volcanic Rocks of the Pilandsberg,” Trans. Geol. Soc.
of South Africa, August 19, 1912; ‘‘The Geology of the Pilandsberg,”’ Annual Report
of the Geol. Survey of South Africa, 1911, p. 77.
756 H. A. BROUWER
in detail. Because the study of the alkali syenites and effusive
rocks which were collected during my visit in 1910 has been post-
poned by my departure for the East Indies, we will mention the
most characteristic features of the latter rocks, as they have been
briefly described by Humphrey.
The richness in pneumatolytic and thermal minerals (especially
fluorspar) is characteristic for the rocks of the whole region; accord-
ingly the loss on ignition is always considerable.
Besides the coarse- to medium-grained intrusive rocks and the
effusive ones, transitional types have a great development. A
series of porphyrites which graduate through all the types between
true lavas, intrusive sheets, and dikes often form an unbroken
series paralleling the effusive rocks.
Nepheline syenites: The nepheline syenites partly belong to the
group of the foyaites and are usually coarse-grained. ‘They are
connected by transitions with the lujaurites, which are characterized
by the abundance of small needle-shaped crystals of aegirine and
are quite similar to the lujaurites of Greenland and the peninsula
of Kola. The latter rocks principally are found in the southern
part of the mountains, while lujauritic rocks only exceptionally
occur in the central part. The foyaites have been frequently found
by Humphrey in dikes, traversing the various effusive rocks and the
red syenites.
If we divide the foyaites according to the character of the dark
minerals, we find that nearly all the subgroups are represented and,
in the porphyritic varieties, we also find aegirine, alkaline amphi-
boles, and biotite, either characterizing different rock types or
occurring together in the same rock. The amphiboles are rich in
alkalies and often show peculiar properties which are similar to
those often mentioned in the literature of the foyaitic and theralitic
rocks. Their optical and chemical properties have not yet been
studied in detail, but they are known to have the properties of
the barkevikitic, kataforitic, or arfvedsonitic amphiboles.
The isolated range of hills to the southeast of the Pilandsberg
proper, which bends around with the lujaurites, consists of aegirine-
amphibole foyaites, in which the amphibole has a pronounced zonal
structure. The differences in color are progressive from brown
ALKALI ROCKS IN THE TRANSVAAL 757
in the central part to green in the margin, while a turning of the
plane of optic axes from parallel to the plane of symmetry in the
central part to normal to it in the margin was often observed.
The extinction angles in sections parallel to [oro] are up to 4o°.
Amphiboles in which the plane of optic axes is normal to the plane
of symmetry have also been found in pegmatitic segregations in
aegirine-amphibole foyaites on Buffelspan [585]. Their angle of
optic axes is very small and the extinction angle b:c is about 14°.
In rocks from Wijdhoek [701] many properties of the amphiboles
agree with those of the green amphiboles which Ussing" described
in rocks from Greenland.
Coarse-grained foyaites are largely developed in the central,
and also in the northern, part of the mountains. The hills and a
part of the valley on and near Boekenhoutfontein [889] consist of
foyaites, which contain aegirine and sometimes are very rich in
biotite; more to the south, at Buffelspan [585], coarse-grained
aegirine-amphibole foyaites are found. Rocks with the same struc-
ture occur in the western part of Houwater [496] and gray foyaites
cover a large surface on Schaapkraal [12]. Leucocratic foyaites
with aegirine as the only dark constituent are found in the western
part of Wijdhoek [7o1], near the eastern boundary of Tusschen-
komst [331]; they are associated with aegirine-amphibole foyaites.
They form a complex of isolated small hills in a valley, surrounded
by ridges of effusive rocks. The foyaites can be followed to the
southern part of Leeuwfontein [429] and to Welgeval [749]. At
Wijdhoek [7o1] they show a considerable amount of variation in
structure and composition; we find gray feldspar rocks, which
contain biotite as the only dark constituent, varieties which are
rich in nepheline, and porphyritic equivalents in which the dark
minerals appear as phenocrysts enclosing the elements of the fine-
grained groundmass.
The lujaurites are characterized by their richness in fine needles
of aegirine. In a forthcoming petrographical paper this group will
be described in detail. Aegirine always predominates; arfved-
tN. V. Ussing, ‘‘Mineralogisk-petrografiske Undersdgelser of Grénlandske
Nefelinsyeniter og beslaegtede Bjaergarter,’’ Meddelelser om Groénland, XIV (1894),
210.
758 H. A. BROUWER
sonite is a rare constituent of the rocks from the Pilandsberg, while
true arfvedsonite lujaurites have been described by Ussing from
Greenland. The lujaurites are connected with the foyaites by
rock types poorer in dark minerals, aegirine with the needle form
gradually disappearing. Like the lujaurites from other regions
(Greenland, Kola peninsula, and Los islands), the rocks of the
Transvaal are characterized by an abundance of rare minerals.
For example, the new mineral molengraaffite,* of the eucolite
group, which sometimes occurs in the foyaites, Is very common in
these lujaurites.
At both sides of the Rustenburg-road, on Ledig [744], we find
tinguaitic rocks and porphyritic lujaurites between the high
lujaurite hills and the norites; to the east the lujaurites can be
followed to Doornhoek [134]. Their southern parts are covered
by effusive rocks. To the northwest, bending round with the
periphery of the Pilandsberg, we can follow the lujaurites to the
southwestern part of Wijdhoek [7or]; farther to the northwest
the periphery of the complex is formed by red to light-red syenitic
rocks. On Tusschenkomst [331] we find the same lujaurites in
contact with the effusive rocks, where a valley separates ridges of
the two rocks.
The tops of the eroded lujaurite ridges are similar to those of
inclined crystalline schists.
We find the lujaurites and their porphyritic equivalents also
in the western, northern, and northeastern part of the complex,
beside the red syenites; in the eastern part, where the effusive
rocks have their greatest development, the latter rocks cover the
whole surface from the periphery to the central foyaites. Probably
the lujaurites occur there at greater depth. The nepheline syenites
and the allied rocks which hitherto have been studied micro-
scopically belong to the following groups:
t. Aegirine foyaites
Leucocratic rocks
Mesocratic rocks
Foyaitic lujaurites
1H. A. Brouwer, ‘‘Molengraaffit, ein neues Mineral in Lujauriten aus Transvaal,’’
Centralbl. f. Min., etc., 1911, p. 120.
ALKALI ROCKS IN THE TRANSVAAL 759
2. Lujaurites
a) Eucolite-molengraaffite lujaurites
b) Eucolite lujaurites
c) Eucolite-astrophyllite lujaurites
d) Aenigmatite lujaurites
e) Lujaurites without eucolite
. Aegirine-amphibole foyaites
. Aegirine-biotite foyaites
. Lujaurite porphyries
. Aegirine-nepheline syenite porphyries without foyaitic structure
. Aegirine-amphibole-biotite-nepheline syenite porphyries
. Tinguaite porphyries
contr OM BW
Syenites: The syenites generally have a red or reddish color;
they principally consist of feldspar, while the dark-colored minerals
have usually been altered into chlorite. Iron ores, titanite,
apatite, and fluorspar are further constituents of these rocks.
Humphrey* mentions that the red syenites of the eastern,
western, and central parts of the Pilandsberg have various points
of dissimilarity in the hand specimens; those of the east, on Rhen-
osterspruit, being almost entirely composed of feldspar, while the
other localities furnish rocks in which is much iron ore. The
latter rocks are very decomposed. The feldspars are microcline,
orthoclase, and anorthoclase. ‘Two analyses of red syenites, which
have been published in the Annual Report of the Geological Survey
of South Africa for 1911, show that there are considerable differ-
ences in chemical composition between the rocks of this group. A
red syenite from Nooitgedacht [748] contains 8 per cent Na,O and
2 per cent KO, while a red syenite from Rhenosterspruit [600]
contains 5 per cent Na,O and 1o per cent K,0. The high potash
content of some of the syenites tends to connect them with the
leucite-bearing effusive rocks, which will be mentioned below.
These red syenites bound the Pilandsberg complex on the north-
western, northern, and southeastern sides. For the greater part
they are developed as a massive wall, forming the outermost circle
of hills at the periphery of the mountains. The red color of the
syenites has given the name to the farm Rooderand [399] and from
tW. A. Humphrey, ‘‘The Volcanic Rocks, etc.,” Trans. Geol. Soc. South Africa,
IgI2, p. 104.
760 H. A. BROUWER
there to Saulspoort [369] the syenites form the periphery of the
complex; to the south, along Ruigehoek [326], Vogelstruisnek
[602], and Palmietfontein [567], they form a nearly interrupted
series of bare, low hills. In the southern part, near the road from
Rustenburg to Saulspoort [269], steep lujaurite hills rise from the
flat norite country. In the southeastern part, where the Rhenoster-
spruit leaves the hills, we again see the bare, red syenite hills on
both sides of the stream. Along the eastern boundary they are
covered by effusive rocks. The syenites are found also in the
central parts of the complex; in the southern part of Driefontein
[888] numerous hills consist of these rocks. They form a conspicu-
ous feature and from a distance can easily be distinguished from the
rounded, bare, felsite ridges. We find them also in the south-
eastern part of Welgeval [749], on Nooitgedacht [748], Buffels-
kloof [219], Leeuwfontein [429], Buffelspan [585], and Houwater
[496]. Near the houses on Nooitgedacht [748], in the valley of a
small rivulet, light syenitic rocks with white feldspars occur, which
are similar to some varieties of the rocks on Leeuwfontein [320] in
the Pretoria district. ;
By transitions these rocks are connected with the nepheline
syenites, as well as with the effusive rocks.
Diorites: As intimately associated with the foyaites and’
lujaurites, Humphrey™ mentions the occurrence of diorites, which
have their greatest development in the northern part of Boeken-
houtfontein [889]. They are also exposed on the summit of the
mountain to the southwest of the native stadt on Saulspoort and
are found as a dike cutting through the norites on Tusschenkomst
[446] to the north-northwest of the Pilandsberg complex. The rock
is fine grained, has a gray color, and consists principally of augite
and labradorite.
Effusive rocks: In my previous paper? it has been stated that _
porphyritic and dense equivalents of the syenites and nepheline
syenites have a great development in the Pilandsberg complex.
Flow structure is often beautifully developed in these rocks.
1“The Geology of the Pilandsberg,” Annual Report of the Geol. Survey of South
Africa, 191I, p. 84.
2 Oorsprong en samenstelling der Transvaalsche Nepheliensyenieten, p. 16.
ALKALI ROCKS IN THE TRANSVAAL 761
The effusive rocks have recently been described in some detail
by Humphrey in a paper on the volcanic rocks of the Pilandsberg.*
He divides the rocks into two main groups—the trachytes and the
phonolites. The first group contains the effusive representatives
of the alkali syenites; the second group, those of the nepheline
syenites. An andesitic rock was found on the ridge separating
the farm Kafferskraal [890] from Saulspoort [269]. It consists of
diallage, diopside, plagioclase, and iron ore in a fine-grained ground-
mass, and may be an effusive representative of the diorites. The
rock has been classed as leucitophyre. The phenocrysts of ortho-
clase are accompanied by phenocrysts of leucite.
The trachytes attain their greatest development in the eastern
portion of the Pilandsberg, on the farms Doornpoort [251] and
Vaalboschlaagte [636], where they measure some 5,000 feet in
-thickness. In this succession the trachytes alternate with red
“‘felsitic”’? rocks and tuffs, while a thick band of leucitophyres
occurs toward the base of the series. These blue-colored leucite-
bearing rocks contain phenocrysts of orthoclase and leucite in a
groundmass of very finely divided aegirine and feldspar. The
phonolites occur in most other parts of the Pilandsberg; they are
of a prevailing greenish and bluish color, contrary to the prevailing
red of the trachytic series. Typical phonolites on the farm Drie-
fontein [888] contain occasional phenocrysts of feldspar in a finely
divided groundmass which consists of feldspar, nepheline, and
much aegirine. In the neighborhood of Saulspoort is a rock con-
taining phenocrysts of sodalite in a cryptocrystalline groundmass.
Volcanic breccias and tuffs are widely distributed throughout
the Pilandsberg.
The effusive rocks of the isolated mountain at the boundary
of Buffelspan [585], Leeuwfontein [429], and Wijdhoek [701] often
have a banded appearance, and a beautiful flow structure with
parallel arrangement of the feldspar phenocrysts is developed.
Well-developed cubes of blue fluorine occur in some of these rocks,
while Humphrey mentions the occurrence of leucite crystals. He
t “The Volcanic Rocks, etc.,”’ Trans. Geol. Soc. South Africa, 1912, p. 105.
_ ?Felsite is a field term under which Transvaal geologists comprise a great diver-
sity of rocks: quartz porphyries, felsites, phonolites, tinguaites, andesites, etc.
762 H. A. BROUWER
found the effusives to form a thick capping resting upon the red
syenites. The effusive rocks are found in the northeastern part
of Buffelspan [585], in the high ridge from Houwater [496] to Wijd-
hoek [7o1], and appearing again at the other side of the Rustenburg
road, where the effusives of the ridge are in contact with lujauritic
rocks and can be followed in a northwesterly direction. On Tus-
schenkomst [331] and Welgeval [749] they are separated from the
lujaurite by a shallow valley at the contact. Still more to the
north we find the effusive rocks on Schaapkraal [12] and on Driefon-
tein [888], where they are exposed in the valley of a rivulet, which
flows in the direction of Rooderand [399]. On the northern farms
of the Pilandsberg the high ridges of effusive rocks bend around
parallel to the circumference of the complex; on the western farms
they have their greatest development and almost entirely hide the
deep-seated rocks.
The rocks of the country around the Pilandsberg.—The rocks
which surround the Pilandsberg complex are the norites and
granites of the Bushveld igneous complex and the quartzites and
shales of the Pretoria series.
Norites and Pyroxenites: These rocks form the characteristic
small hills (Pyramids, Zwartkoppies) parallel to the Magaliesberg
range. They bend to the northwest in the neighborhood of
Rustenburg, but the characteristic hills disappear long before they
reach the Pilandsberg; much more to the north, on the farm
Modderkuil [565], we see them again just in the continuation of
those to the south of the Pilandsberg. The bands of magnetite
are found to the southeast of the Pilandsberg. ‘They end against
the red syenites near the boundary of the farms Rhenosterfontein
[867] and Rhenosterspruit [906], but are found again to the north
of the Pilandsberg. We see that the whole southern part of the
Pilandsberg is immediately surrounded by the basic rocks; on
the farms Ledig [744] and Koedoesfontein [818] they are in imme-
diate contact with lujaurites and allied rocks. Near the boundary
of Zandrivierspoort [747] and Mahobieskraal [562] the isolated
hills of aegirine-amphibole foyaites and the ridges of Magaliesberg
quartzite come close together. At a small distance farther to the
northwest and to the west the basic margin of the Bushveld com-
ALKALI ROCKS IN THE TRANSVAAL 763
plex is again largely developed. It covered the whole region,
which is limited to the south by the Schurvebergen and Zeerust
Hills, to the west by the Marico Hills, and to the north by the
Dwarsbergen.*
In following the basic margin along the western boundary of
the Pilandsberg, we find quartzites in the northern part of Vogel-
struisnek [602] which are in immediate contact with red syenites.
More to the north the latter rocks border again upon norites.
Near, and west of, the native stadi on Ruigehoek [426] norites rich
- in feldspars are exposed in the valley of a rivulet; they are the same
rocks as those which are found to the south of the Pilandsberg, but
show a pronounced schistose structure and a dip to the northeast.
To the west the rocks become more basic, and near the contact
with the quartzites on Davidskuil [142] very basic rocks were
collected in the valley of a rivulet which is crossed by the road from
the native stadt on Mabieskraal [620] to Janskop on Bierkraal [545].
They, too, show a pronounced schistose structure. Here the strike
is about N. 15 W., and the dip is to the east-northeast.
. On Tusschenkomst [446], to the east of the quartzite hill Janskop
on Bierkraal [545], a series of hills consisting of schistose basic
rocks can be followed in a north-northwesterly direction. Hum-
phrey? mentions a peculiar feature of the pyroxenites, particularly
noticeable on the farms Ruigehoek [426] and Zandspruit [181],
where narrow bands of chromite, dipping to the east, have formed a
band of comparatively high ground and an apparent stratification.
From all that has been said above, it is evident that the basic
margin of the plutonic complex is cut off abruptly by the intrusion
of the Pilandsberg. :
Granites: The red granites of the igneous complex of the Bush-
veld are found to the east of the norites. The boundary between
the two rock types crosses the Elands River in the southeastern
part of Rhenosterfontein [867] and ends against the alkali syenites.
The red granites are found all along the eastern part of the
Pilandsberg; on Saulspoort [269], west of the Rustenburg road, the
«F, H. Hatch, Trans. Geol. Soc. South Africa, VII (1904), Dp. I.
2“The Geology of the Pilandsberg,”’ Annual Report of the Geol. Survey of South
Africa, 1911, p. 81:
764 H. A. BROUWER
boundary between the norites and granites begins against the
effusive rocks of the Pilandsberg complex and runs from there in
a northeasterly direction. ‘The occurrence of brecciated rocks with
granite boulders in the hill behind the Saulspoort Mission station, :
which was already mentioned in my previous paper, has been
studied in detail by Humphrey,’ who found various types of igne-
ous rocks. The relationship between these is very complicated.
Syenite is seen to be intrusive into the effusive rocks and fragments
of granite are found within the syenites and effusive rocks. Farther
up the hill there is an extensive outcrop of granite which extends
for some 800 yards along the face of the hill. Above this granite
is found a diorite, and the crest of the hill is formed by effusive
rocks. Breccias, in which granite occurs as included boulders,
and also repeated outcrops of granite were found on Doornpoort
[251] and Zuiverfontein [718] in the eastern marginal part of the
Pilandsberg. Large boulders of red granite embedded in coarse
red syenite are to be seen in the bed of the Rhenosterspruit on the
farm Rhenosterspruit [609].
Pretoria series: The Magaliesberg Range, which from Rusten-
burg strikes in a northeasterly direction, comes to an abrupt end
on Mahobieskraal [567], to the southeast of the Pilandsberg com-
plex. Then the Pretoria beds bend to the west; near Bechuanaland
they have a short northerly direction, and then return again
to the east, passing at a distance of about 8 miles to the north
of the Pilandsberg complex and forming the northern boundary of
the igneous complex of the Bushveld.
Isolated hills of quartzite are found at several places to the east
of the Pilandsberg. On Vogelstruisnek [602] they are in immediate
contact with the red syenites. Other hills of quartzite occur on
Tweelaagte [180], Vlakfontein [902], behind the native stadt on
Mabieskraal [620], on Davidskuil [142], and still more to the north
on Bierkraal [545]. From Janskop on Bierkraal the quartzite hills
extend still more to the east, where they approach the northern
boundary of the igneous complex of the Bushveld.
Between the Pilandsberg and the Marico River, the Upper
Magaliesberg beds are missing from the normal sequence of the
Pretoria series. They are represented by the isolated hills of
«The Geology of the Pilandsberg,”’ Annual Report of the Geol. Survey of South
Africa, 1911, p. 87.
ALKALI ROCKS IN THE TRANSVAAL 765
quartzite, which are entirely surrounded by rocks belonging to the
igneous complex of the Bushveld and most probably were broken
up in connection with the intrusion of this complex.
Dike rocks.—The first nepheline syenite of the Transvaal was
discovered by Elie Cohen’ in 1872, near the Hex River, between
Renseburg and Rustenburg. He states that this rock forms the
lower parts of the Zwartkoppies, where these hills bend to the
northwest. The rocks collected by Cohen were described by E. A.
Wiilfing? in 1886 as porphyritic foyaites in which the nepheline
only occurs in the groundmass.
In 1904 G. A. F. Molengraaff* collected nepheline syenites on
the farms Elandsheuvel [255] and Tweede Poort [189]; these rocks
are porphyritic foyaites in which nepheline occurs as phenocrysts.
These rocks are described in the petrographical part of my previous
paper. Since that time the sheet Rustenburg (sheet No. 4) has
been mapped by the Geological Survey of the Transvaal, and it is
shown that several dikes of these porphyritic foyaites intersect
the rocks of the Bushveld complex, running from the Pilandsberg
in a southeasterly direction.
Some of them even cut through the Magaliesberg Range to the
east of Rustenburg and can be followed still farther to the south.
A fine-grained red syenitic rock was found by me to the north
of the red hill on Rooderand [399], cutting through the norites in a
nearly northerly direction. At the boundary of Groenfontein [302]
and Bierkraal [545] near the quartzites, a porphyritic syenite was
found in the basic rocks. On Plate XIV in the annual report for
ro1r of the Geological Survey of South Africa it is shown that
several syenitic dike rocks can be followed from the Pilandsberg
in a north-northwesterly and northwesterly direction.
All these dikes cut through the norites and granites of the
Bushveld, and they have been intruded after the consolidation of the
rocks of the Bushveld igneous complex.
= E. Cohen in Berichte tiber die XVI. Versammlung des Oberrheinischen Geologischen
Vereins, am 29 Marz, 1883, Stuttgart.
2E. A. Wiilfing, “‘Untersuchung eines Nephelinsyenits aus dem mittleren Trans-
vaal,” Neues Jahrb. f. Min. Geol. u. Pal., 7 Mai, 1888, Bd. I, p. 16.
3G. A. F. Molengraaff, ‘‘Preliminary Note on the Geology of the Pilandsberg,”
Trans. Geol. Soc. South Africa, VIII (1905), 208.
766 H. A. BROUWER
Inside the Pilandsberg several dikes occur. Some dike rocks
with the macroscopic appearance of the tinguaites form a band of
comparatively high ground, because of their resistance to denuding
agencies. On Boekenhoutfontein [889] near the boundary with
Kafferskraal [890] a dike of this kind strikes N. 50 W.; it measures
about 10 meters across and cuts through the foyaites with biotite;
the contact with the foyaites is formed by a bent line, as well as
the contact of a dike in the southeastern part of Koedoesfontein
[649].
Near the boundary of Driefontein [889] and Nooitgedacht [148]
I found a tinguaitic dike, 2 meters in diameter, with a sharp contact
and a northeasterly strike, cutting through medium-grained
nepheline syenites in the valley of a rivulet. Near the contact
the rock has a glassy appearance; in the central part the structure
is porphyritic. This dike dips steeply: to the northeast. At the
boundary of Olivenfontein [145] and Rooderand [398], in the valley
to the south of the red syenites, a similar dike, which measures 5
meters across, is exposed.
The direction of these dikes agrees nearly with that of the dikes
outside the Pilandsberg. In the western rivulet to the north of the
houses on Driefontein [888], a tinguaitic dike or segregation,
averaging 40 centimeters in width, has a blended contact with the
surrounding lujaurites. It is rich in bronze-brown flakes of mica.
Near the lujaurites the rock is very rich in aegirine; this mineral
is often developed in spherulites which are up to 1 centimeter in
diameter.
According to Humphrey,’ dikes of foyaite, red syenite, nepheline
syenite, and diorite occur in all parts of the Pilandsberg, and, in
addition to these, there are many basaltic and tinguaitic varieties
occurring in various parts. In the spruit on Saulspoort [269] a
dike of red syenite cuts through the effusive rocks. A series of red
“‘felsitic”’? dikes and blue-black glassy dikes, which were difficult
of determination, and dikes of nepheline syenite traverse the red
syenites. The dikes of nepheline syenite, which have their greatest
development outside the Pilandsberg, seem to disappear into the
«““The Geology of the Pilandsberg,” Annual Report of the Geol. Survey of South
Africa, 1911, p. 85.
ALKALI ROCKS IN THE TRANSVAAL 767
nepheline syenites in the central part of the complex. Dikes of
foyaite were found traversing the red syenites of the central part.
From all that has been said above, it is evident that the foyaites
are the youngest rocks of the Pilandsberg complex, while there is
good evidence to show that the red syenite is older than some of the
effusive rocks.
Pegmatiies.—Dikes of pegmatite, which in many other nepheline
syenite regions contain numerous rare minerals, were not met with
during my visit. Coarse-grained pegmatitic segregations in the
normal-grained rocks are of frequent occurrence, but the rare
minerals were not found in much larger crystals than in the normal-
grained varieties.
Pegmatites rich in eucolite are well exposed in lujauritic rocks
from the hills to the north of the houses on Driefontein [888]. They
consist principally of large crystals of feldspar, green nepheline,
long crystals of aegirine, and carmine-red eucolite which is partly
altered to catapleiite; they also contain some astrophyllite. The
prisms of aegirine are up to ro centimeters in length, and some-
times show a graphic intergrowth with feldspar. Near, and to the
west of, the main road to Saulspoort [269] where it crosses the
Rhenosterspruit in the northeastern part of Buffelspan [585], we
found pegmatites in the aegirine-amphibole foyaites. Feldspars
up to ro centimeters in length, green nepheline, prisms of amphi-
bole, and prisms or spherulites of aegirine are the main constituents;
they also contain some fluorine. Amphiboles with a very small
angle of optic axes im which the plane of the optic axes is normal to
the plane of symmetry’ occur in these rocks. In the southern part
of Wijdhoek [701], near, and to the west of, the main road and to
the south of the ridge of effusive rocks, we found pegmatites, which
are very rich in astrophyllite and spherulites of aegirine, measuring
up to several centimeters in diameter. They are found still
farther to the southwest on Koedoesfontein [746] in a rivulet which
joins the Wolvespruit, where numerous blocks of pegmatites and
lujaurites could be collected; some of them are very rich in eucolite.
=H. A. Brouwer, ‘On Zonal Amphiboles in Which the Plane of Optic Axes of the
Margin Is Normal to That of the Central Part,” Proceed. Kon. Akad. Amsterdam, XVI
(1913), 275.
768 H. A. BROUWER
The aegirine-biotite foyaites in the northwestern part of Boeken-
houtfontein [889] contain pegmatitic segregations. and small dikes
in which feldspars, feldspathoids, and eucolite mineral and small
pale-yellow needles occur. Often they are very rich in aegirine;
this agrees with its tardy crystallization in most of the rocks of
the region.
Finally, we found segregations in the lujaurites to the west of
the common boundary post of Tusschenkomst [331], Leeuwfontein
[429], and Wijdhoek [7or]; they consist almost entirely of aegirine
spherulites which are up to some centimeters in diameter.
Segregations rich in fluorine occur in the microfoyaites of
Olivenfontein [745], and segregations rich in large aegirine spheru-
lites occur in mesocratic foyaites of the valley, running in a north- |
south direction in the northeastern part of Buffelspan [585].
Humphrey mentions the occurrence of very coarse-grained
pegmatites with much fluorspar on Doornhoek [134] and beautiful
pegmatites, about half a mile from the homestead on Driefontein
[888] on the main road to Buffelskloof (219).
Mechanism of intrusion of the Pilandsberg complex.—Since the
rocks of the Pilandsberg complex are younger than the red granites
and norites of the Bushveld igneous complex, and since the Pilands-
berg is surrounded on three sides by norites and on one side by
red granites, it seems to be beyond doubt that the space which is
occupied by the Pilandsberg intrusive rocks was occupied, prior
to the intrusion, by the norites and red granites of the Bushveld.
That the removal of the original rocks was not the result of
folding is proved by the occurrence of a great number of vertical
dikes of vast extension, which are genetically connected with the
intrusion.
The hypothesis that the subsidence of crust blocks elsewhere
was the cause of the intrusion of the magma and the hypothesis
of laccolithic intrusion seem not to be applicable in the present case.
As has been stated by Humphrey, there can be no doubt that
the Pilandsberg represents the remnant of what was once an impor-
tant focus of eruption, and the hypothesis that the intrusive magma
has filled up the cavities which were formed by volcanic outbursts
of an explosive character seems to be applicable.
ALKALI ROCKS IN THE TRANSVAAL 769
Tuffs and volcanic breccias are found all over the areas where
the effusive rocks are developed. The main rock types of the
Pilandsberg are disposed in concentric circles, of which the outer-
most consists of syenites and nepheline syenites and is followed
toward the center by a ring of effusive rocks. The latter dip, with
a few local exceptions, from the center outward, and the highest
hills formed by intrusive rocks in the central area of the complex
still carry a capping of volcanic rocks which have resisted denuda-
tion.
If the Pilandsberg is considered as the remnant of what was
once a volcano and its subsidiary peripheral vents, this must have
been of stupendous dimensions, since the intrusive rocks cover a
surface whose diameter varies from 15 to 18 miles, the lavas having
extended far beyond the periphery of the intrusives.
It is peculiar that in the territory of the Pilandsberg effusive
rocks are found in large quantity between the granular rocks,
whereas they do not occur in the surrounding granites and norites.
The lavas, which must have extended far beyond its periphery,
have entirely disappeared and do not even cap the hills of Magalies-
berg quartzite, though at some places the quartzite is found in the
immediate neighborhood of the Pilandsberg. It is very likely
that in connection with the intrusion of the alkali rocks the roof
has locally sunk down, and, while it has disappeared everywhere
else in the neighborhood by erosion, we see the remains preserved
just on those spots where the roof has given way.
Subsidences in ancient volcanic regions are by no means rare.
Judd,’ for instance, mentions the comparatively perfect state of
preservation exhibited by the great volcano of Mull, if compared
with that of the other great Tertiary volcanoes in the Hebrides.
It can be shown that this difference is due to a central subsidence
which took place in the Mull volcano. From the sections along the
shores of the deep fiords it is evident that the basaltic lava sheets
dip toward the central mass of eruptive rocks, the inclination
increasing as we approach the volcano. Further, there is clear
tJ. W. Judd, ‘‘On the Ancient Volcanoes of the Highlands and the Relations of
Their Products to the Mesozoic Strata,” Quart. Journal of the Geol. Soc., XXX (1874),
250.
770 H. A. BROUWER
evidence of the existence of faults, the downthrow of which is in
all cases toward the great central mass. A similar subsidence
took place after the period of the eruption of acid lavas and before
that of the basaltic lavas.
The state of preservation of the Pilandsberg complex and sur-
rounding area is not very favorable to a study of the amount of
subsidence in the sunken area. The lavas, which must have
extended far beyond the mountain proper, have entirely disap-
peared; the junction of the intrusive rocks of the Pilandsberg with
those of the Bushveld is not well exposed; and the amount of
denudation in the area surrounding the complex cannot be esti-
mated. This probable subsidence and the large dimensions of
the plutonic body lead us to mention another hypothesis to explain
the mechanism of intrusion of many batholites, which has been
set forth by Daly.‘ He termed this process “overhead stoping”’;
it consists of a continued breaking free of roof blocks and a sinking
down of the detached blocks into the magma, which consequently
rises and occupies the place of the sunken fragments.
The cover of the intrusive rocks of the Pilandsberg entirely
consists of lavas, the effusive equivalents of the intrusive rocks,
and this is very common in batholitic intrusions from other parts
of the world. How these facts are explained by overhead stoping
has been elaborately discussed by Ussing? in a recent treatise on the
geology of the country around Julianehaab. If a batholitic magma
on one or more occasions during its intrusion has penetrated its
cover, this will presumably lead to a volcanic outburst of cata-
strophic character, accompanied by the outpouring of lava flows
and followed by a period of quiescence. After a time, when hot
magma from below is brought into contact with the newly formed
roof, the stoping process will continue, interrupted by few volcanic
outbursts until the magma has cooled to its point of solidification.
In several batholites with a permanent cover of sedimentary
rocks the stoping process came to an end and the magma was
R. A. Daly, ‘Geology of the Ascutney Mountain,” Un. St. Geol. Surv. Bull.
No. 209 (1903), p. 93; ‘‘The Mechanics of Igneous Intrusion,” Amer. Jour. of
Science, 4th series, XV (1903), and XXVI (1908).
2N. V. Ussing, ‘‘ Geology of the Country around Julianehaab, Greenland,” Medd.
om Grénland, XXXVIII (1911), p. 302.
ALKALI ROCKS IN THE TRANSVAAL Te
solidified before the earth’s surface was reached, but the Pilands-
berg alkali magma must have been very rich in mineralizing agents,
which reduced its viscosity, and in such magmas the stoping process
may go on when they near the earth surface and until the cover is
penetrated. Volcanic outbursts cause an escape of the volatile
substances, and the magma becomes more and more viscous, until
a new supply of heat and mineralizers from below sets up stoping
again. In fact, several rare minerals with a highly complex con-
stitution, which are not stable at high magmatic temperatures,
occur within the Pilandsberg rocks. Fluorine is a very common
constituent, and, as their very name imports, fluorides must have
reduced the viscosity considerably. Moreover, fluorine and other
minerals in which we find direct evidence of the co-operation of
mineralizers are regularly distributed in several rocks of this region,
where they crystallized in the last cavities, thus proving that the
mineralizing agents in part were regularly distributed until the
final consolidation.
Of course, direct support would be given to the co-operation of
overhead stoping if fragments which could only be derived from
an original cover of the crystalline rocks were found among the
rocks of the Pilandsberg complex, but Humphrey' mentions that
all the close-grained rocks, which in the hand specimens very much
resemble shales, proved themselves under the microscope to be
devitrified lavas.? Particularly at those places in the northeastern
part of the area where the granites are found to within a few hun-
dred yards of the Pilandsberg complex it is of great interest to
know whether these granites occur in their original position.
Age of the Pilandsberg.—In the neighborhood of the Pilandsberg
the rocks which formed the covering of the igneous complex of the
Bushveld at the time of its intrusion most probably belonged to the
t “The Volcanic Rocks of the Pilandsberg, etc.,” Trans. Geol. Soc. South Africa,
IQI2, p. 102.
2TIn my previous paper (Oosprong en samenstelling der Transvaalsche Nephelien-
syenieten), p. 17, I mentioned having found shales in the valley to the north of
the homestead on Houwater [496], but the rocks were not studied under the
microscope.
3 Cf. also H. A. Brouwer, ‘‘On the Formation of Primary Parallel Structure in
Lujaurites,” Proc. Kon. Akad. Amsterdam, 1912, p. 734.
772 H. A. BROUWER
Waterberg system; they have here been entirely removed by denu-
dation. Humphrey mentions that there are no signs of the presence
of Waterberg rocks among the stratified lavas, nor were any frag-
ments of those rocks found among the volcanic breccias, while
many examples of included granite boulders within the Pilandsberg
rocks were found. :
He concludes ‘that the volcanic outbursts and the outpouring
of lava postdated the removal of all of the sedimentaries of the
Waterberg system in this neighborhood. No evidence is available
about the age of the Pilandsberg rocks with regard to the Karroo
system.
Of course, if the possibility of overhead stoping is admitted, the
problem of the age of the Pilandsberg rocks is more complicated,
but the question of the mechanism of intrusion is still too vague
for further discussion.
OTHER OCCURRENCES OF NFPHELINE SYENITES AND ALLIED ROCKS
The occurrence of nepheline syenites on Leeuwfontein [320] and
Zeekoegat [287] was discovered by Molengraaff in 1898. The
numerous variations of the Leeuwfontein foyaites in chemical and
mineralogical composition and also the leucocratic and melano-
cratic dike rocks, bostonites, monchiquites, tinguaites, etc., were
described. Liebenerite porphyries, like those which occur at
Predazzo in the Tyrol and at Alné (Sweden), are also associated with
the nepheline syenites of this region.
D. Draper discovered nepheline syenites on Walmansdal [116]
to the northwest of Zeekoegat [287]. The rocks to which J. A. L.
Henderson? gave the name hatherlite were also collected on Leeuw-
fontein [320]. As was stated by Molengraaff,? the name hatherlite
is not applicable because the old powder factory “ Eerste fabrieken”
or “Hatherley factory”’ is situated to the south of the Magaliesberg
Range and has nothing to do with the factory on Leeuwfontein [320].
1G. A. F. Molengraaff, ‘‘Note on Our Present Knowledge of the Occurrence of
Nepheline Syenite in the Transvaal,” Trans. Geol. Soc. South Africa, VI (1903), p. 89.
2 J. A. L. Henderson, On Certain Transvaal Norites, Gabbros, and Pyroxenites and
Other South-African Rocks, London, 1808.
3G. A. F. Molengraaff, Geology of the Transvaal (Johannesburg, 1904), p. 46.
ALKALI ROCKS IN THE TRANSVAAL 77.2
In the Annual Report of the Geological Survey of the Transvaal
for 1903, A. L. Hall’ gives an account of the rocks from Leeuw-
fontein [320] and Zeekoegat [287] which is followed by a petro-
graphical description of these rocks and those on Walmansdal [116]
and the newly discovered occurrence on Franspoort [426].
On Leeuwkrall [396], about 5 km. to the northwest of Hamans-
kraal station, H. Kynaston? discovered two occurrences of syenitic
rocks, the southern one locally graduating into nepheline syenite.
It is a porphyritic foyaite similar to some of the dike rocks which
occur to the southeast of the Pilandsberg. On Rietfontein [451]
and Spitskop [463] an interesting occurrence of nepheline syenites
within the red granites was discovered by Hall’ in 1910; and
Wagner’ mentions the occurrence of a dike of basic camptonite
cutting through the Waterberg sandstones on Buffelspruit [1920],
which means probably that nepheline syenites occur at a deeper
level.
The intrusion on Leeuwfontein [320].—To the east of Pretoria,
near Franspoort [426], the ridges of quartzite belonging to the
Pretoria series bend to the southeast; the Magaliesberg quartzites
have been extended in length, while the Daspoort and Timeball
quartzites were strongly pressed in a direction slightly oblique to
the strike of the strata.
In describing the dislocations connected with the intrusion of
the igneous complex of the Bushveld, Molengraafi> supposed that
at those places where the circumference of the complex shows a
convex curve interesting phenomena may be expected. We saw
that the Pilandsberg intrusion is located where the Pretoria series,
tA. L. Hall, ‘‘On the Area to the North of the Magaliesberg Range and to the
East of the Pietersburg Railway Line,” Annual Report of the Geol. Survey of the Trans-
vaal, 1903, p. 38.
2H. Kynaston, “‘On the Area Lying North-West of Pretoria, between the Maga-
liesberg Range and the Salt Pan,” Annual Report of the Geol. Survey of the Transvaal,
1905, p. 29.
3 Annual Report of the Geol. Survey of the Transvaal, 1910.
4P. A. Wagner, ‘‘Note on an Interesting Dyke Intrusion in the Upper Waterberg
System,” Trans. Geol. Soc. South Africa, 1912.
5G. A. F. Molengraaff, Proc. Geol. Soc. of South Africa, 1905; “Criticism on
Messrs. A. L. Hall and F. A. Steart: On Folding and Faulting in the Pretoria Series,”
Trans. Geol. Soc. South Africa, VIIL (1905), 7-15.
774. H. A. BROUWER
which from Rustenburg strike in a northeasterly direction, bend
again to the west, and the nepheline syenite intrusions to the north-
east of Pretoria are found where the ridges of quartzite bend to the
southeast. The foyaite intrusion on Franspoort [426] is entirely
surrounded by Magaliesberg quartzite and is clearly intrusive in
them. Near the intrusion on Leeuwfontein [320], which borders
‘“‘felsites’’ only on the north, the Magaliesberg quartzites cover a
large surface in consequence of numerous faults. Following the
valley of the Pienaars River to the north, we see a succession of
red “‘felsites’’ with an approximate east-west strike and a varying
dip to the north, alternating with eruptive breccias, conglomerates,
and basic effusive and dike rocks. More to the north, on Roode-
plaat [314], they are covered by shales and syenitic rocks of doubtful
age, and on Paarderfontein [338], at a great distance from the
foyaite, dikes of tinguaitic and andesitic character, which in part
are connected with the intrusion on Leeuwfontein [329], cut through
the sandstones and conglomerates of the Waterberg system. The
chemical composition of a tinguaite of Paardenfontein [338] closely
agrees with that of the normal foyaites on Leeuwfontein [320].
The small differences are similar to those which characterize the
nepheline syenites and accompanying tinguaite dikes from other
regions.
An interesting dike of basic camptonite has recently been
described by Wagner.t It occurs on Buffelspruit [1920] in the
Waterberg district and cuts through Waterberg sandstone. This
proves again that the intrusion of nepheline syenite with which
the dike most probably is connected is younger than the Waterberg
sandstones.
The ‘‘felsites’’ are the effusive equivalents of the intrusive
rocks on Leeuwfontein [320]. The liebenerite porphyries, which in
the southern part of Roodeplaat are exposed over a long distance
in the valley of the Pienaars River, show the same characteristics
as the liebenerite porphyries of Alnéd and the Tyrol. But also the
dense weathered rocks of which the mineralogical composition
could not be recognized under the microscope belong to the alkali
»P. A. Wagner, ‘‘ Note on an Interesting Dyke Intrusion in the Upper Waterberg
System,” Trans. Geol. Soc. South Africa, 1912.
ALKALI ROCKS IN THE TRANSVAAL 775
rocks, as was proved by chemical tests. After treating the powder
with hydrofluoric acid, only 0.4 of its weight was evaporated, and
a simple calculation makes evident that the Al partly occurs in
feldspars, partly in feldspathoids. Microchemically the residue
gave a strong soda reaction and a very feeble potash reaction.
From this it is evident that ‘‘felsites’? which are the effusive
equivalents of the intrusive rocks are genetically connected with
the alkali rocks of the intrusion on Leeuwfontein [320].
An exact petrographical examination will greatly assist in the
determination of the stratigraphical place of the different ‘‘felsites.”’
Identity in age for the felsophyres of the Waterberg district and
of the phonolites of Leeuwfontein [320] would seem to be in the
highest degree improbable.
From a petrographical point of view there is much resemblance
between the rocks of Leeuwfontein [320] and those of the Pilands-
berg; the association of foyaites of varying composition with red
syenites and effusive rocks is a common characteristic. The rocks
on Leeuwfontein [320] near the old dynamite factory are principally
red syenites and red hololeucocratic feldspar rocks; in the south-
ern part the leeuwfonteinites with accompanying porphyritic equiva-
lents occur. The porphyritic rocks sometimes form well-defined
dikes.t Leeuwfonteinite porphyry and monzonite porphyries are
found between Leeuwfontein [320] and Franspoort [426] and along
the path to Derde Poort [469]. The numerous varieties of foyaite
occur near the boundary of the farms Leeuwfontein [320] and
Zeekoegat [287]; they will be described in detail in a forthcoming
petrographical paper. The normal foyaite of this region is a coarse-
grained, leucocratic, aegirine-amphibole foyaite. In varieties
rich in feldspathoids (particularly sodalite) aegirine is the only dark
constituent; they pass into rocks which are nearly free from feld-
spar (¢awites). Rocks very rich in titanite (pienaarites) occur at
several places. The rocks on Leeuwfontein [320] differ from those
of the Pilandsberg by the absence of rare minerals in the latter
rocks. In other nepheline-syenite regions the rare minerals are
1 The leeuwfonteinites are the same rocks as Henderson’s hatherlites (anortho-
clase syenites), cf. Henderson, On Certain Transvaal Norites, Gabbros, and Pyroxenites
and Other South-African Rocks. They contain much plagioclase and their composition
varies between that of the alkali monzonites and that of the alkali syenites.
776 H. A. BROUWER
also often limited to the aegirine foyaites and the arfvedsonite
foyaites and are wanting in the foyaites with barkevikitic amphi-
bole. The rare minerals and a not very small quantity of lime in
the magma seem to exclude one another (compare the analysis of
the normal foyaite of Leeuwfontein, which has been given in my
previous paper). The association of foyaites with leeuwfonteinites
which besides barkevikite also contain plagioclase makes it
probable that the CaO content of the common mother-magma was
rather considerable.
The rocks of the neighborhood of Leeuwfontein [320], which
hitherto have been studied under the microscope, belong to the
following groups:
1. Aegirine foyaites
Leucocratic rocks
Pienaarites (melanocratic rocks
rich in titanite)
. Aegirine-amphibole foyaites
. Tawites
. Feldspar rocks
. Aegirine-foyaite porphyries
. Aegirine-amphibole foyaite por-
phyries
. Leeuwfonteinites
. Leeuwfonteinite porphyry and
monzonite porphyry
9. Tinguaite porphyries
10. Monchiquites
11. Augitites
12. Andesitic camptonites
13. Doleritic nepheline basalts
14. Diabases
15. Liebenerite porphyries
16. Bostonites
17. Phonolites
Aun fhW bd
comt
H. Kynaston' mentions that the foyaite of Walmansdal [116] is
clearly intrusive in the ‘‘felsites.”
Nepheline syenite region to the west of Lydenburg.2—This region
covers a surface which has about the same extension as that on
Leeuwfontein [320]. It is surrounded by red granites and occurs
H. Kynaston, “The Geology of the Country Surrounding Pretoria,” Explanation
Sheet I, Geol. Surv. of the Transvaal, 1907, p. 28.
2 A.L. Hall, in Annual Report Geol. Surv. of the Transvaal, 1910.
ALKALI ROCKS IN THE TRANSVAAL 777
close behind the zone of ultra-acid rocks which Hall discovered at
the boundary between the granites and the basic margin of the
igneous complex of the Bushveld. The specimens which Mr. Hall
kindly put at my disposal during my stay in the Transvaal are
melanocratic lujaurites and lujaurite porphyries, which sometimes
show a schistose structure. The colorless minerals are sometimes
very subordinate and microscopically the rocks seem to consist
almost wholly of fine needle-shaped crystals of aegirine. These very
melanocratic lujaurites were rare in the Pilandsberg complex, but
seem to cover the greater part of this newly discovered occurrence.
Foyaites also occur, and it is interesting to find the association
of lujaurites with leucocratic feldspathoid rocks (urtites), which
consist chiefly of nepheline. The association of lujaurites and
urtites in the peninsula of Kola (they received their names from
the same place—Lujavr Urt) is also a characteristic of this district.
An isolated mass of strongly metamorphic limestone is inclosed
within the alkali rocks.
ORIGIN AND AGE OF THE NEPHELINE SYENITES AND ALLIED ROCKS
It does not seem improbable that the nepheline syenites have
originated from the same sources as the granites and norites of the
Bushveld. The formation of the basic margin in the main intru-
sion of the Bushveld proves that magmatic differentiation took
place on a very large scale. ‘Toward the periphery the rocks become
more and more basic, while granites occupy the central portion.
When tested in detail, the view of general increase of basicity from
the center toward the periphery requires modification. Hall" has
described a zone of ultra-acid rocks with 97 per cent SiO, in the red
granites close to the boundary with the norites to the west of Lyden-
burg. He considers these rocks as a product of extreme differen-
tiation, which could take place near the basic margin, when the
viscosity of the granitic magma was already strongly increased.
That sometimes the acid and basic rocks pass gradually into one
another possibly depends on the depth to which the complex has
been exposed by erosion.
1 A. L. Hall, ‘Note on Certain Widespread Ultra-Acid Rocks Occurring along the
Margin of the Bushveld Granite, etc.,”’ Trans. Geol. Soc. South Africa, XIII (1910),
ps) 1O:
778 H. A. BROUWER
If ultra-acid rocks have differentiated from the granitic magma,
the residual magma will be enriched in Al,O,, and alkalies with
regard to SiO, and its composition will more or less agree with
that of the nepheline syenites. This kind of differentiation may
have taken place on a much larger scale at a greater depth, which
has not been exposed by denudation. It is interesting to find an
occurrence of nepheline syenites on Rietfontein [451] and Spitskop
[463], close behind the zone of ultra-acid rocks.
The coarse textures, the rather indefinite order of crystallization,
the numerous poikilitic structures,’ the abundance of fluorine and
rare minerals with a highly complex constitution, which are char-
acteristic for many of the nepheline syenites in the Transvaal,
make it probable that these rocks crystallized from a residual
magma in which the volatile constituents were concentrated and
which may have crystallized at rather low temperatures. Some
of the nepheline syenites are certainly younger and may be con-
siderably younger than the sandstones and conglomerates of the
Waterberg system; the time at which the different intrusions have
risen to the present level and the time at which they have con-
solidated may have varied between wide limits.
The age of an intrusive rock is determined by the time of its
consolidation, and it is very probable that the alkali magmas
remained fluid during a very long period of igneous activity.
When these magmas which are rich in volatile substances shall
crystallize will greatly depend upon the eventual loss of these
substances, which may have been the immediate cause of crystal-
lization quite as much as of any actual cooling.’
Only some characteristic features of the various igneous rocks
have been dealt with; as has been stated above, it does not seem
improbable that the nepheline syenites and allied rocks have
originated from the same sources as the granites and the norites
of the Bushveld. ‘To the petrologist there are many very interest-
ing problems with regard to the origin and age of the different
rock types which would repay further research.
tH. A. Brouwer, ‘‘On Peculiar Sieve Structures in Igneous Rocks Rich in Alka-
lies,” Proc. Kon. Akad. Amsterdam, November, to1t.
2 A. Harker, The Natural History of Igneous Rocks, 1900, p. 186.
PETROLOGICAL ABSTRACTS AND REVIEWS
ALBERT JOHANNSEN
FLETT, J.S.,and Hi11, J.B. The Geology of the Lizard and Meneage.
Mem. Geol. Surv., Sheet 359. London, 1912. Pp. 280, pls.
15, figs. 10, bibliography 10 pp.
The rocks of the Lizard, probably of Archean age, represent an
igneous complex of serpentine, gabbro, and gneiss, surrounded by an
aureole of hornblende-schists and metamorphosed sedimentary rocks—
mica-schists, green-schists, and quartz-granulites. The hornblende-
schists were originally basic igneous rocks, but they are now so much
altered that their original character as extrusive, tuff, or intrusive cannot
be determined in every case. Some of them contain sedimentary mate-
rial and possibly represent volcanic ashes. The time that elapsed
between the formation of the schists and the intrusion of the serpentine
is not known, but most of the rocks of the aureole probably were already
in a metamorphosed condition at the time of the intrusion of the basic
rock. Besides the serpentine there is also a coarse hornblende-schist
in some places, which may represent dolerite sills intruded immediately
before the basic rock.
Numerous chemical analyses and photogravures of thin sections
make the memoir a valuable work for reference. Jt is a notable con-
tribution to the literature of serpentine.
Frett, J. S. ‘The Geology of the Lizard,” Proc. Geologists’
Assoc., XXIV (1913), 118-33, pls. 3, map I.
A brief summary of the preceding paper on the geology of the Lizard,
intended for the use of members of the Geologists’ Association on their
Easter excursion, 1913.
Fiett, J. S., and Hitt, J. B. ‘Report of an Excursion to the
Lizard, Cornwall,’ Proc. Geologists’ Assoc., XXIV (1913),
313-27, pls. 4.
Foye, Witsur G. ‘‘Nephelite-Syenites of Haliburton County,
Ontario,” Amer. Jour. Sci., XL (1915), 413-36, figs. 9.
The nephelite-syenite laccoliths of Haliburton County, central
Ontario, are described together with the associated rocks. A number
779
780 PETROLOGICAL ABSTRACTS AND REVIEWS
of analyses with recomputations in the C.I.P.W. system are given.
Among the rocks are syenite, canadite, nephelite-pegmatite, various
contact rocks, hornblende-nephelite rock, monmouthite, biotite-nephelite
rock, and pegmatitic nephelite-syenite. The writer thinks that the close
association of the granite-pegmatite with the nephelite-syenite indicates
that they originated from a primary granite magma at about the same
time. Following Daly, he thinks that the nephelite-syenites. were pro-
duced by the action of limestone on the granite magma.
GoLtpMAN, Marcus I. ‘“Petrographic Evidence on the Origin
of the Catahoula Sandstone of Texas,” Amer. Jour. Sci.,
XXXIX (1915), 261-87, figs. 12.
Thinks the Catahoula sandstone originated from wind-blown sand
in an arid region. The arrangement of fossils indicates subaérial burial
in blown sand in some cases, and burial by wind, but in a quiet body of
water, in others. Evidence for the interpretation of disintegrated sedi-
ments in general is considered in detail.
Kato, Takeo. ‘Mineralization in the Contact Metamorphic
Ore Deposits of the Ofuku Mine, Prov. Nagato, Japan,”
Jour. Geol. Soc. Tokyo, XX (1913), 13-32, pl. 2, figs. 3.
The copper ores of the Ofuku mine are contact metamorphic deposits
in sedimentary rocks at a short distance from an igneous body. They
are accompanied by typical contact minerals, such as wollastonite,
garnet, vesuvianite, etc., which were deposited metasomatically from
solutions derived from the igneous magma. The character of the solu-
tions changed gradually during the period of metamorphism. At first
they were very siliceous; later they became more basic, and rich in iron
and silica and with more or less sulphide ores, and finally very basic and
rich in copper and iron sulphide and poor in silica.
Koto, B. “On the Volcanoes of Japan,” Jour. Geol. Soc. Tokyo,
XXIII (1916), 1-13, 17-28, 29-55, to be continued.
These three papers represent the beginning of a series of articles by
Doctor Koto on the Japanese volcanoes. Of the 170 post-Tertiary
volcanoes of Japan, 55 are active. All the recent lava is andesitic, but
some of the earlier flows were plagioliparite and basalt. The writer
describes each volcano in brief form, classifying the cones according to
the system proposed by Schneider, and gives references to previous work.
PETROLOGICAL ABSTRACTS AND REVIEWS 781
Kozu, S. ‘Petrological Notes on the Igneous Rocks of the Oki
Islands,” Science Repts. Tokoku Imp. Univ., Sendai, Japan.
Second Series, Vol. I (1913), No. 3, 25-56, pls. 4, figs. 5.
The Oki Islands lie about 65 kilometers off the coast of Honshu,
on the Korean side. They consist mainly of volcanic rocks extruded
between the middle of the Tertiary and the beginning of the Pleistocene,
and lie upon or were intruded into the Tertiary beds which form the base
of the islands. The succession of the igneous rocks cannot be exactly
determined in all cases but it appears to be as follows, beginning with
the most recent:
4
Oo
. Holocene sediment, a river deposit of limited extent in valleys
. Trachydolerites and basalts
. Pleistocene (?) deposits
. Trachytic rocks
. Trachydolerites
. Banded alkalic rhyolites
. Alkalic rhyolites
. Quartz-syenites and schistose granitic rocks
. Andesites
. Tertiary deposits
HdnHW Po AN CMO
The various rock-types are described in detail, chemical analyses
are given for most of them, and the names in the C.I.P.W. system are
determined.
REVIEWS
Physiography of the Beaverdell Map-Area and the Southern Part of
the Interior Plateaus of British Columbia. By LEopoLp
REINECKE. Geol. Surv., Canada, Museum Bull. No. 11,
1O05. /Ppw5s, pls.s5) es. 3, map-1-
A study of the Beaverdell area is of value because its history is
characteristic of the whole plateau region of British Columbia. Follow-
ing the eruption of lavas (Nipple Mt. series) in early Miocene a mature
topography was developed. Late in the Tertiary the canyon-cutting
stage was inaugurated by an uplift of about 1,000 feet. Pliestocene
glaciation, first by continental ice from the north and later by valley
glaciers, modified the youthful topography. The present areal ratio of
uplands to valleys is three to one.
Ho RoR:
Report on the Copper Deposits of the Eastern Townships of the
Province of Quebec. By J. AusTEN BANcRorT. Dept. of
Colonization, Mines and Fisheries, Mines Branch, 1915.
Pp. 295, pls. 10, figs. 9, map tf.
Lenticular bodies of pyrite carrying a little chalcopyrite occur in
highly metamorphosed igneous and sedimentary rocks. A large pro-
portion of the deposits have been formed by irregular impregnation
and replacement along shear zones within altered igneous rocks. Other
deposits occur at the contacts of the intrusives and as impregnations
or partial replacements of limestones. The schistose intrusives seem
to have been the source of the sulphide ores.
Development in this region was begun during the Civil War, when
the price of copper was abnormally high. Since 1869 only a few com-
panies have operated. Four properties have yielded large profits; no
others have repaid the money spent upon them. The future develop-
ment of the mines depends largely on the utilization of the sulphur
content of the ores in the manufacture of sulphuric acid, or other
chemicals.
H.R. B.
REVIEWS 783
Notes on the Geology and Paleontology of the Lower Saskatchewan
River Valley. By E. M. Kinpte. Geol. Sur., Canada,
Museum Bull. No. 21, 1915. Pp. 25, pls. 4.
Description of Silurian sections and faunas, including new species,
Leptaena sinuosus and L. Parvula.
RB.
The Geology and Mineral Resources of the Buller-Mokthinui Sub-
division, Westport Division, New Zealand. By P. C. Morcan
and J. A. Bartrum. New Zealand Dept. of Mines, Geol.
Surv., Bull. No. 17, new series, 1915. Pp. 210, pls. 109, figs. 1,
maps 9.
This area is situated on the northeast coast of the South Island of
New Zealand. The rocks are described as consisting of the Aorere
series of metamorphosed Siluro-Ordovician sediments intruded by pre-
Triassic granites, a coal-bearing Eocene series, the Oamaru series of
Miocene age, and Quaternary deposits, both Pleistocene and Recent.
The Westport district is famous for its gold placers, fluvial and
marine gravels having yielded a total of £4,675,000. The industry has
greatly declined in recent years. The Eocene coal is a high-grade
bituminous variety. The total tonnage is estimated at 123,000,000
tons, of which 60,000,000 is extractable. ‘The Miocene series contains
considerable quantities of brown and lignitic coal.
HER. 3B:
The Squantum Tuillite. By RoBert W. Sayers. Bull. Mus.
Comp. Anat., Harvard College, LVI, No. 2 (1914), 141-75,
pls, 12
For many years the origin of the Roxbury conglomerate has been a
subject of debate. As early as 1875 W. W. Dodge stated his belief in
the glacial origin of these beds; the writer has at last established this
view. The Roxbury series, comprising the Roxbury conglomerate, the
Squantum tillite, and the Cambridge slate, is of late Paleozoic age,
probably Permian. If there is no duplication of beds by folding, the
tillite is 600 feet thick. It is an unstratified mass of unassorted materials
much affected by dynamic movements, with the development of sec-
ondary cleavage. The rock fragments are of several kinds, variable in
size, and mostly angular or subangular in shape. Striated stones were
found at four localities.
784 REVIEWS
The character of the included rock fragments suggests that the ice
moved from the southeast. Intercalated slate beds indicate recessions
of the ice. Whether they represent temporary retreats or long inter-
glacial epochs is not known. Two conglomeratic beds below the prin-
cipal tillite are of probable glacial origin, though it is not certain that
they were deposited directly by the ice. Heaps
Geology of the Lake Pleasant Quadrangle, Hamilion Co., N.Y. By
Wiiiiam J. Mitter. New York State Museum, Bull. No.
162, TOrO:\ Eps, pis. fo, figs. 4 map ts
The Lake Pleasant Quadrangle lies in the south central Adirondacks.
The Grenville series of meta-sediments and intrusives outcrops over
most of the region and is cut by a network of normal faults. Two
small areas of Paleozoic strata are preserved by the dropping of fault
blocks. The maximum thickness of this section is 500 feet. The
formations preserved are: Potsdam sandstone, Theresa beds, Little
Falls dolomite, Black River (Lowville) limestone, Trenton limestone,
and Canajoharie (Trenton) shale.
The normal syenites of the region grade into basic syenites, also into
granitic syenites and granites. The basic phases are attributed to the
assimilation of dark Grenville gneisses. Pure differentiation has been
the principal factor in the production of the silicic phases. ‘Transitions
from gabbro into basic syenite are described as due to assimilation by
the gabbro. Hopp:
Geology and Underground Waters of the Northern Llano Estacado.
By Cuartes L. Baker. Bull. Univ. Texas, No. 57, 1915.
Pp. 225, pls. 10, maps 3.
For half a century or more the Llano Estacado has been famous for
its stock-raising. Recently there has been a serious attempt to utilize
the ground water for purposes of irrigation. The supply of shallow
water is found to be insufficient to irrigate all the land that it underlies.
Conservation is therefore of first importance, but unless dry farming
proves more successful than in the past this region will always be chiefly
a stockman’s country.
Previous geologic work is largely confirmed by the present study.
The strata represented are the Permian red-beds, the Upper Triassic
Dockum group, comprising the Tecovas and Trujillo formations, marine
beds of upper Comanchean age, possibly some Cretaceous rocks, and
imperfectly known Miocene and later Cenozoics. Hook By
REVIEWS 785
Petroleum and Natural Gas in Oklahoma. By C. W. SHANNON
and L. E. Trout. Okla. Geol. Surv., Bull. No. 19, Part I,
HOUSE, Vepsr3e.hplse7, fies. A.
The great public demand for information concerning oil and gas
necessitates the publication, not only of detailed reports on individual
fields, but also of papers containing general information regarding the
oil and gas business. The present bulletin is designed to meet both
demands. It is published in two parts. Part I deals with the general
phases of the industry, and includes a short discussion of the geology of
Oklahoma. Part II gives a more detailed account by counties of the
oil and gas fields of the state. HiRue.
The Willow Creek District, Alaska. By S. R. Capps. U.S. Geol.
Surv., Bull. No. 607, 1915. Pp. 86, pls. 15, figs. 5.
An area of 90 square miles at the head of the Little Susitna River,
which enters Cook Inlet from the north, is described. The geologic
formations are pre-Jurassic mica schists cut by quartzdiorites, and an
Eocene sedimentary series with interbedded basaltic lava flows. Alaskite
dikes and gabbro masses occur in association with the larger intrusives.
Gold occurs in placers and quartz lodes. The latter are quartz-filled
fissures in the quartz-diorite carrying free gold and sulphides.
Hi oRe eB:
The Broad Pass Region, Alaska. By FRrepD A. Morrit. U.S.
Geol. Surv., Bull. No. 608, 1915. Pp. 80, pls. 8, figs. 3.
The Broad Pass region comprises an area of about 3,700 square miles
along and south of the axis of the Alaska Range east of Mount McKinley.
The oldest rocks are of Devonian age, representing the same general
horizon as the Devonian of the Mount McKinley and Porcupine River
regions. The Mesozoics are less deformed than the Devonian series.
Basic lava flows are apparently overlain by upper Triassic slates which
are probably equivalent to the “undifferentiated Paleozoic” series of
slates and graywackes along the south flank of the Alaska Range, as
described by Brooks, Capps, and Eldridge. A series of slate, graywacke,
and conglomerate is provisionally assigned to the Jurassic. At some
time before the Tertiary these rocks were folded and intruded by igneous
masses. The Eocene is represented by the Cantwell formation, in
places intensely folded and,cut by granites and diorites. The sections
on Quaternary deposits, igneous rocks, and glaciation are by Joseph
E. Pogue. H.R. B.
786 REVIEWS
Geology and Oil Resources of the West Border of the San Joaquin
Valley North of Coalinga, California. By ROBERT ANDERSON
and RoBERT W. Pack. U.S. Geol. Surv., Bull. No. 603, 1915.
Pp. 220, pls. 14, figs. 5. .
The region described in this bulletin is a strip 8 to 20 miles wide and
130 miles long lying along the east flank of the Diablo Range and the
adjacent western edge of the San Joaquin plain. This foothill belt was
studied with a view to determining whether or not oil fields exist, as in
the Coalinga and other districts farther south. In so far as finding
new oil fields is concerned, the examination proved disappointing.
Organic shales were found in great quantity, but there are few spur
folds running out toward the valley. The presence of oil pools to the
south is determined by such folds.
The oldest rocks in the Diablo Range belong to the Franciscan
formation, of supposed Jurassic age, which is separated from the oldest
Cretaceous rocks by a great unconformity. The Knoxville group
(Comanche) is believed to be absent. The Chico (Cretaceous) is
represented by the Panoche and Moreno formations, marine shales and
sandstones that reach the enormous thickness of 24,000 feet. The
Moreno, composed largely of organic remains such as diatoms and
foraminifers, has been referred by some authors to the Tertiary, but it
is now found to contain Cretaceous fossils. Hitherto strata of this type
have been known in California only in the Tertiary.
The Martinez (lower Eocene) is present only in the southern part of
the area, where it is represented by 5,000 feet of marine beds. The
Tejon (upper Eocene) is present throughout the region, varying in
thickness from 50 to 2,200 feet. The Oligocene is represented by the
Kreyenhagen diatomaceous shale, with unconformities above and below.
In the southern part of this belt the Miocene is represented, as in the
Coalinga district, by the Vaqueros, Santa Margarita, and Etchegoin-
Jacolitos formations, with a maximum thickness of 5,000 feet, each
separated from the next adjoining by an unconformity. The Big Blue
serpentine, formerly considered to represent the lower part of the
Santa Margarita, contains typical Vaqueros fossils. Farther north the
lower and middle Miocene were not differentiated. The San Pablo is
equivalent in part at least to the Etchegoin-Jacolitos. Post-Miocene
beds up to 2,200 feet thick are tentatively correlated with the Tulare
formation. ;
The local factors influencing the accumulation of oil, evidences of
oil in the region, and the future possibilities of development are dis-
REVIEWS 787
cussed in detail. The oil is believed to have been derived from the two
organic shales, and apparently each gave rise to a different type of
oil—the Moreno to a light paraffin oil and the Kreyenhagen to a heavy
asphalt oil. The diatoms are believed to have been the greatest contri-
butors in the formation of the oil.
ERe Be
Mineral Resources of Alaska for t914. By Atrrep H. Brooks
and Others. U.S. Geol. Surv., Bull. No. 622, 1915. Pp. 380,
pls. 11, figs. 8.
This volume is the eleventh of a series of annual bulletins summariz-
ing the results of the investigations of Alaskan mineral resources and
the status of the industry in the territory. Fourteen papers deal with
the mineral resources of certain districts.
The gold and copper deposits of the Port Valdez district are descibed
by B. L. Johnson. The country rock includes basic lavas, slates, gray-
wackes, and other sediments of Mesozoic age. Gold occurs in quartz-
filled fissure veins formed at moderate depths; the copper chiefly as
sulphide impregnations and replacements of sheared zones along the
fractures. The mineral association in both gold and silver ores is
practically the same, varying only in relative proportions. The sulphide
minerals are pyrite, chalcopyrite, galena, sphalerite, and some pyr-
rhotite and arsenopyrite. ‘There was but one period of mineralization.
As in the Ellamar district, both types had a common origin in solutions
that circulated subsequent to late Mesozoic intrusions, with which they
were probably genetically related.
P. S. Smith and A. G. Maddren describe the quicksilver prospects
of the Kuskokwin region. The ore occurs in brecciated zones in Cre-
taceous sandstones and shales at the contacts of granitic and andesite
dikes. Cinnabar, generally with stibnite, occurs in quartz veinlets
and stringers. In places calcite and siderite are present. Some cinnabar
has also been obtained from placer gravels, and detritus in a stream near
one of the deposits contains native mercury.
Ee gReE Bs
RECENT PUBLICATIONS
—WnuiTBEcK, R. H. The Geography of the Fox-Winnebago Valley. [Wis-
consin Geological and Natural History Survey. Madison, 1915.]
—WituAms, M. Y. An Eurypterid Horizon in the Niagara Formation of
Ontario. [Canada Department of Mines, Museum Bulletin No. 20,
No. 1574, Geological Survey, Geological Series, No. 29. Ottawa, October
8, 1915.]
Arisaig-Antigonish District, Nova Scotia. [Canada Department of
Mines, Memoir 60, No. 1398, Geological Survey, Geological Series, No. 47.
Ottawa, 1914.]
—WILLISTON, S. W. (1) The Osteology of Some American Permian Verte-
brates.. II. (2) Synopsis of the American Permo-Carboniferous Tetra-
poda. Contributions from Walker Museum, Vol. I, No. 9. [Chicago:
The University of Chicago Press, 1916.]
—Woop, Harry O. Effects in Mokuaweoweo of the Eruption of 1914.
[American Journal of Science, 4th Series, Vol. XLI, No. 245, May, 1916.]
—Woopwarp, H. P. The Reputed Petroliferous Area of the Warren River
District (Southwest Division). With a Contribution on Asphaltum from
the Southern Coast of Australia, by E. S. Stmpson, and a Petrological
Contribution by R. A. FarquHarson. [Geological Survey of Western
Australia, Bulletin 65. Perth, ro15.]
788
EN DEX LO \ZOLUME XX
Abele, Charles A. Statistics of the Mineral Production of Alabama for
1913. Review by A. D. B. ;
Active Volcanoes of New Zealand, The. By E. 9: WMieore ‘
Adams, Frank D., and J. Austen Bancroft. On the Amount of itavetiall
Friction Developed in Rocks during Deformation and on the
Relative Plasticity of Different Types of Rocks.
Adirondack Intrusives. By H. P. Cushing
Adirondack Intrusives. By N. L. Bowen . Soe on ne
Adirondacks, : Structure of the Anorthosite Bada in che: By, Hee:
Cushing ,
Age and Relations of the ial Baan Remains ound at avers Florida’
Symposium on the .
Alkali Rocks in the Transvaal, On the Cooley of che Be H. A.
Brouwer : 5
American Institute of Mining Engineers, Transactions of the. Vol. L.
Review by A. D. B.
Ibid., Vols. XLVIII, XLIX. Reviews Bye D. B.
Aicerican Mining Congress, Proceedings of the. Revere by A D. B.
Amorphous Minerals, A Review of the. By Austin F. Rogers
Anderson, Robert, and Robert W. Pack. Geology and Oil Resources
of the West Border of the San Joaquin Valley north of Coalinga,
California. Review by H. R. B.
Anorthosite Body in the airen dacs) erderare ai He: By H. P.
Cushing
Anorthosites, The Problem of che! By N. its Bowen :
Arisaig-Antigonish District, Nova Scotia. Py M. Y. Walle Re-
view by W.B.W. .
Atwood, Wallace W. mother Nvocakey of ioeene (Glacation in
Southern Colorado . :
Eocene Glacial Deposits of Soniean Colorado: Revie
by H.R. B. :
Baker, Charles L. Geology and Underground Waters of the Northern
Llano Estacado. Review by H. R. B. Sa ae
Bancroft, J. Austen. Report on the Copper Deposits of the Hastert
Townships of the Province of Quebec. Review by H.R. B. 5
Bancroft, J. Austen, Frank D. Adams and. On the Amount of Internal
Friction Developed in Rocks during Deformation and on the
Relative Plasticity of Different Types of Rocks. :
Barlow, A. E. Corundum, Its Occurrence, Distribution, Eeplonation
and Uses. Review by AY Ds Be HEAR ata
789
790 INDEX TO VOLUME XXV
Barrell, Jorenl Central Connecticut in the Geologic Past. Review by
W. B. W.
Beaverdell Map-Area aad the Souther Part of he Tatenee Platewus of
British Columbia, Physiography of the. By Leopold Reinecke.
Review by H. R. B.
Beekly, A. L. Geology and Coal Resources OF Nord Bork Colorado:
Review by C. W. T. :
Berg, Georg. Die Tak voacopicche Untersuchung aes rvingerstiittent
Review by A. D. B. 2 ;
Berry, Edward W. The Fossil Plants fron Ver. Blonde aera
Boise, C. W. Diamond Fields of German South-West Africa. Review
by CW ra ; : ‘
Boone County [W. Va.]. By G. E. Kvens nadeDy) D. Teta. ie Review
by WalB Wei:
Bowen, C. F. The SHrigra phy of We Nontana Group rath Special
Reference to the Position and Age of the Judith River Formation in
North-Central Montana. Review by V. O. T.
Bowen, N. L. Adirondack Intrusives
The Problem of the Anorthosites .
Bowie, William. Our Present Knowledge of Isostasy front Geodeue
Evidence
Breccias, A Classification ae Shales th: Sradents i W. H. Nonton
Bretz, J Harlen. The Satsop Formation of Oregon and Washington
Bridge, Josiah. A Study of the Faunas of the Residual Mississippian of
Phelps County (Central Ozark Region), Missouri Din eel as
Brines in Manitoba, The Corrosive Action of Certain. By R. C.
Wallace : MI Ma Tey Wy case fieceeti roca
prone Hass Region, iis ‘The. By Fred A. Moffit. Review by
Bae shes SE gk
Brooks, A. H., and Others Mineral Recounees of lac Review ay
Weary nh
Mineral Resources ¢ of ade och IQI4. Revie by H. R. B.
Brouwer, H. A. On the Geology of the Alkali Rocks in the Transvaal.
Buehler, H. A. Biennial Report of Missouri Bureau of anes and
Mines. Review by W. B. W.
Buller-Mokihinui Subdivision, Westport Dison! New ealand. The
Geology and Mineral Resources of the. By Boi. Morgan and
J. A. Bartrum. Review by H. R. B.
Burchard, Ernest F. The Red Iron Ores of East Memneseee! tegen
by We O. T.
Burling, Lancaster D. Dowawarning: its Toine Blanes at #he Glose
of the Niagaran and Acadian. PER ahr tol!
Cairnes, D. D. Upper White River District, Yukon. Review by
W. B. W. Rene:
Calkins, F. C. A Becmal Grouping of tHe Blaeioclacce
Campbell, M. R. Glacier National Park. Review by W. B. wW.
Canadian Geological Survey, Summary Report of. Review by W. B. W.
INDEX TO VOLUME XXV
Canyon Range, West-Central Utah, A Reconnaissance in the. By G. F.
Loughlin. Review by V. O. T.
“‘Capillarity, Some Effects of, on Oil NGaraiitan ee a) A. W. McCoy,
Discussion of. By C. W. Washburne
Capps, Stephen R. The Willow Creek District, see Review by
RB:
The Mentae Disence: Aleabee Review by iL R. B.
Central Connecticut in the Geologic Past. By Lee Barrell. tase
DYE Bo Wee cial),
Chamberlin, Rollin T. Fucker Shudies at Vv ero, Hlecidee
Interpretation of the Formations Pane Human Bones
at Vero, Florida
Champlain Sea in the Lake Gaerne Pati The. 1 Rareley, F. Mather:
Clay and Shale Deposits of the Province of Quebec, Rattan, Report
onthe. By J) Keele. Review by’ A.D. Bs. *
Coal Fields of Kittitas oan eer By ails euunderes Revicu
yas. Wis: h.
Coal Fields of Pierce lcauney By focepns Tnicles Renee by
W. B. W.
Coal Measures, The Westen Tnerce Kecosicinc afl Its Bearing on
the Origin and Distribution of the. By Francis M. Van Tuyl.
Coalville Coal Field, Utah, The. By Carroll H. ceca Review
lon KOON bara
Colorado Ferberite and the Wolfen ite Series: By 1 De Be Hex and
W. T. Schaller. Review by W. B. W.
Common Minerals and Rocks: Their Occurrence a Wee a R. D.
George. Review by V. O. T.
Copper Deposits of the Eastern Townships of fe Province i Ouebee
Report onthe. By J. Austen Bancroft. Review by H.R. B.
Corrosive Action of Certain Brines in Manitoba, The. By R. G:
Wallace
Corundum, Its eenmence: DICHBUCOAE BS ploieation and cee By
A. E. Barlow. Review by A. D. B.
Cretaceous-Eocene Contact in the Atlantic and Gulf onstall ita
The. By L. W. Stephenson. Review by V. O. T. :
Cretaceous Formations, Relations of, to the Rocky Mountains in Col-
orado and New Mexico. By WillisT. Lee. Review by H.R. B.
Cross, Whitman, and Esper S. Larsen. Contributions to the as
raphy of Southwestern Colorado. Review by V. O. T.
Cushing, H. P. Adirondack Intrusives
Structure of the Anorthosite Body in me Adinondacles
Cuzco Valley, Peru, A Geologic Reconnaissance of the. By Herbert E.
Grevonyca Reve wilyy Car Ws Mee icc OE es unas ireeg | fr ce, Wile a eae
Dales a. N elson, and Others. Slate in the United States. Review by
W. B. W.
Daniels, Joseph. Coal Fields of Bienes Cone Review by W. B. W.
400
792 INDEX TO VOLUME XXV
Darton, N. H., and Others. Guidebook of the Western United States.
Part Cc The Santa Fe Route. Review byi@ Way !
Davis, W. W. . Evidence Bearing on a Possible Northeastward Baten!
sion of Mississippian Seas in Illinois ; :
Decimal Grouping of the Plagioclases, A. By F. C. Venting :
Devils Lake, Wisconsin, The History of. By Arthur C. Trowbridge
Devonian of Southwestern Ontario. By C. R. Stauffer. Review by
W. B. W. ;
Diamond Fields of Geran: South: West Africas By C W. Boies
Review by C. W. T. :
Dip Protractor, A Proposed. By Ghee K. Wentworth
Discussion of ‘Some Effects of Capillarity on Oil Accumulation,” ae
A. W. McCoy. By C. W. Washburne
Downwarping along Joint Planes at the Close of ane Nearan ond
Acadian. By Lancaster D. Burling :
Eakin, Henry M. The Iditarod-Ruby Region, Alaska. Review by
Voom Pe MRR AAU sen ee tee
Elastico-Viscous Flow, The lane ae By A. A. Michelson :
Eocene Glacial Deposits of Southwestern Colorado. By Wallace W.
Atwood. Review by H.R. B. ‘
Eocene Glaciation in Southern Colorado, Another Locality AE By
Wallace W. Atwood. ~ :
Eruption of Mauna Loa, Notes on the 1916, I ond If. By eieaey O.
Wood . : : Dea CAG ous
Ibid. TIL and IV. HSS aa rae
Eruption of Usu, A Few Interesting phenomena on fhe: By Y. Oinouye
Evidence Bearing on a Possible Northeastward Extension of Mississip-
pian Seas in Illinois. By W. W. Davis :
Faunas of the Residual Mississippian of Phelps County (Central Ozark
Region), Missouri, A Study of the. By Josiah Bridge
Fettke, Charles Reinhard. The Manhattan Schist of Ghucheastern
New York State and Its Associated Igneous Rocks. Review By
Albert Johannsen
Flett, J.S. The Geology of the Te ara Review be Alber: jonenncese
Flett, J. S., and J. B. Hill. Report of an Excursion to the Lizard,
Cornwall. Review by Albert Johannsen ; é : ,
The Geology of the Lizard and Meneage. Review by Albert
Johannsen : See Pm
Foerste, August F. Infraformational Bepples 7 fhe Richmond Group
at Winchester, Ohio. ; ! ; : é ;
Foothills Structure in Northern Calas iByy Victor Ziegler. }
Fossil Bird Palaeochenéides Mioceanus, The Relationships of the. By
Alexander Wetmore : : : : :
Fossil Plants from Vero, Florida, The | By Edward W. Berry :
Foye, Wilbur G. Nephelite-Syenites of Haliburton oe Ontario.
Review by Albert Johannsen
PAGE
399
576
157
344
402
400
489
584
145
199
405
687
684
322
467
258
576
558
503
779
779
779
280
715
555
661
779
INDEX TO VOLUME XXV
Gasoline and Benzene-Toluene, The Manufacture of, from Petroleum
and Other Hydrocarbons. By W. F. Rittman, CTs. Dutton, and
E. W. Dean. Review by A. D. B.
Geologic Reconnaissance of the Cuzco Valley, Ben A. Be Herbert E.
Gregory. Review by C. W. T.
Geological Relations and Some Fossils of eeuih Georan By i W.
Grecory.. wweview byl Ee) Re Bis,
Geology and Coal Resources of North Ron Colorado! By A. io
Beekly. Review by C. W. T.
Geology and Mineral Resources of the Buller: Mokininns Subdinston!
Westport Division, New Zealand, The. By P. C. Morgan and
J. A. Bartrum. Review by H. R. 1 : Hina
Geology and Oil Prospects of Northwestern Oresens By C. W. Wash-
burne. Review by W. B. W.
Geology and Oil Resources of the West Border of the San Toran Valley
north of Coalinga, California. By Robert Anderson and Robert W.
Pack. Review by H. R. B.
Geology and Paleontology of the Lower Gaskatchenan River Valley,
Notesonthe. By E.M. Kindle. Review by H.R. B.
Geology and Underground Waters of the Northern Llano er eador
By Charles L. Baker. Review by H.R. B. ‘
Geology and Water Resources of Tularosa Basin, New Mexico. By
O. E. Meinzer and R. F. Hare. Review by C. War.
Geology. of the Alkali Rocks in the Transvaal, On the. By H. A.
Brouwer ' : :
Geology of the Hanagita- Bremner Region of Alaska, By F. H. Moffitt.
Review by W.B.W. _. AN be
Geology of the Lake Pleasant Quadrangle, Hamilton Co, N.Y. By
William J. Miller’ Review by H.R. B.. :
Geology of the Lizard and Meneage, The. oe Joon blett ond i. B. ‘Hill.
Review by Albert Johannsen ;
Geology, of the Wizard, The. By J.-S. “Flett. Revie iy Albert
Johannsen : : : ‘
George, R. D. Common Minerals and Rocks: Their Occurrence and
Uses. Review by V. O. T. 5 Se a a RAT
Glacial Deposits of Southwestern Colorado, Bocese: By Wallace W.
Atwood. Review by H.R. B. Mr yeaa MAUS sa
Glacier National Park. By M.R. Cuanall Iain by W. B. W.
Gold-Bearing Gravels of Beauce Co., Quebec, The. By J. B. Tyrrell.
Review by A. D. B.
Goldman, Marcus I. BeeroE phic Evidence! on he Origin oi the
Catahoula Sandstone of Texas. Review by Albert Johannsen .
Gold on the North Saskatchewan River. By J. B. Tyrrell. Review
bya DB:
Gold-Platinum- Palladium Lode in 1 Southern Nevada, A. By Adolph
Knopf. Review by A. D. B.
Grabau, Amadeus W. Age and Sr eraple ieelinene of the Ole
tangy Shale of Central Ohio, with Remarks on the Prout Limestone
and So-called Olentangy Shales of Northern Ohio é ;
794 INDEX TO VOLUME XXV
Gregory, Herbert E. A Geologic Reconnaissance of the Cuzco Valley,
Peru. Review by C. W. T.
Gregory, J. W. Geological Relations Ane Some. Fossils lok South
Georgia. Review by H. R. B. ‘
Guidebook of the Western United States. Part B, The ‘Overland
Route. By W. T. Lee, R. W. Stone, H. S. Gale, and Others.
Review by C. W. T. j : f : ; } :
Guidebook of the Western United Spates Part C, The Santa Fe Route.
By N. H. Darton and Others. Review by C. W. T.
Gypsum Deposits of the Maritime Provinces. By William F. Fennison
Review by W. B. W. :
Hanagita-Bremner Region of Alaska, Coleey of the. ey 1 goal ale
Moffit. Review by W.B.W. .
Hancock, E. T. The History of a Potont of Varga River Colors.
and ts Possible Bearing in That of Green River. Review by
Hay, Oliver P. The Ouatcuiary: ecocits at Vero lord and the
Vertebrate Remains Contained Therein
Hennen, R. V. Wyoming and McDowell Counties Iw. Var Review
by WeBoWei
Hennen, R. V., and D. B. Reon Moon andl Mingo Counties [W. a. L
Review by W.B. W. ..
Hess, F. L., and W. T. Schaller. Calowide herbonite and the Woolies
Series. Review by W. B. W. :
Hill, James M. Some Mining Districts in Noriheasvern Galacns ana
"Northwestern Nevada. Review by C. W. T
Hinds, Henry, and F. C. Greene. Stratigraphy of the Pennsylvanian
Series in Missouri. Review by W. B. W. .
Hotchkiss, W. O., and Edward Steidtman. amectone Road Materials
- of Wisconsin, Review by W. B. W.
Hrdli¢ka, Ales. Preliminary Report on Finds of Supposedly Ancient
Human Remains at Vero, Florida. :
Hudson Bay Basins and Upper Mississippi River. U. S. Geol: Survey
Water-Supply Paper, 355. Review by W.B. W. .
Huels, F. W. Peat Resources of Wisconsin. Review by W. B. W.
Human Bones at Vero, Florida, Interpretation of the Formations Con-
taining. By Rollin T. Chamberlin
Human Remains and Artifacts at Vero, Florida, Nate on mine Deposits
Containing. By E. H. Sellards :
Human Remains and Extinct Vertebrates at ere! lord! On the
Association of. By E. H. Sellards
Human Remains at Vero, Florida, On Reporred Pletoccnes By
Thomas Wayland Vaughan
Human Remains at Vero, Florida, erent Report on hinds of
Supposedly Ancient. "By Ales Hrdli¢ka
Human Remains Found at Vero, Florida, Symposium on ihe Age and
Relations of the Fossil Se RC Snes
PAGE
. 306
596
397
399
I0o
40
43
INDEX TO VOLUME XXV 795
PAGE
Iditarod-Ruby Region, Alaska, The. By Henry M. Eakin. Review
by) V- ©..T. 199
Igneous Rocks, Suceestions fon a Quamtirative Naneralepieal Classifica:
tion of. By Albert Johannsen. . 63
Internal Friction and Limiting Strength of Rocke ander Gonticon: of
Stress Existing in the Interior of the Earth, On the Mathematical
Theory of the. By Louis Vessot King Sin iA 638
Internal Friction, On the Amount of, Developed in hoa ahve
Deformation ‘and on the Relative Plasticity of Different Types of
Rocks. By Frank D. Adams and J. Austen Bancroft ; ESO,
Intraformational Pebbles in the Richmond cue: at Winchester, Ohio.
By August F. Foerste. : 250
Isostasy, On the Hypothesis of. Be, W. D. MacMillan a He TOS
Isostasy, Our Present Knowledge of, from Geodetic Buidences By
William Bowie . : 3 : y } Ri hile
Japan, On the Volcanoes of. By B. Koto. Review by Albert Johann-
sen ; : : t 780
Jaw of the tou Man, The. By Getret $. Miller, Jr. Reviow or
EM REABY: eo 506
Jennison, William F. Copaginn epee i the Wartime Provinces
Review by W.B.W._ . : inere)
Johannsen, Albert. Petrological becrets an Rerens. dae AO) ear 779
Review of Mineralization in the Contact Metamorphic Ore
Deposits of the Ofuku Mine, Prov. Nagato, Japan, by Takeo
Keaton: : 780
———. Review of Nephelite: epeniies of Haliburton Conmes Oileit.
byaWalbur GeFoye : AS
———. Review of On the Volcanoes ai Tapan by B. Koto i j Noo
———. Review of Petrographic Evidence on the Origin of the Cata-
houla Sandstone of Texas, by MarcusI. Goldman .. 780 »
———.. Review of Petrological Notes on the ie Rocks af the
Oki Islands, by S. Kozu 781
Review of Report of an Excursion to ihe aed: Connell
‘by J. S. Flett and\J. B. Hill. Nein SM airKo)
———. Review of The Geology of the Lizard, ay 1 5. F lett. eMC MG
Review of The Geology of the Lizard and Meneage, by J. S.
-Flett and J. B. Hill : 779
———. Review of The emer Schist of Souchewetera New Vor
State and Its Associated Igneous Rocks, by Charles Reinhard Fettke 593
Suggestions for a Quantitative Mineralogical Classification of
Igneous Rocks. . . Pay aes 63
Jones, Fayette A. Mineral Resources ie INGE Mexico! Review by
aN LDN oe ie : 3 t : i : ; ; 5 BxOK0)
Kato, Takeo. Mineralization in the Contact Metamorphic Ore
Deposits of the Ofuku Mine, Prov. Raat, te aa Review by
Albert Johannsen . Fe AE SO
796 INDEX TO VOLUME XXV
Keele, J. Preliminary Report on the Clay and Shale Deposits of the
Province of Quebec. Review by A. D. B.
Kindle, E. M. Notes on the Geology and paleoneolons of the lanes
Saskatchewan River Valley. Review byH.R.B. .
Some Factors Affecting the Development of Vide @ racks
King, Louis Vessot. On the Mathematical Theory of the Internal Fric-
tion and Limiting Strength of Rocks under Conditions of Stress
Existing in the Interior of the Earth : ; ;
Knopf, Adolph. A Gold-Platinum-Palladium iade Mia Southern
Nevada. Review by A. D. B. t 4 : s
Koto, B. On the Volcanoes of Japan. Reviey, by Albert Johannsen .
Kozu, S. Petrological Notes on the Igneous Rocks of the Oki Islands.
Review by Albert Johannsen Laie HA Oe eat a a
Krebs, C. E., and D. D. Teets, ae Boone County [W. Va.] . Review by
WoB won Pia SLR eae dns li
Labidosaurus Cope, a Lower Permian eect Reptile from Texas.
By Samuel W. Williston
Lake Pleasant Quadrangle, Hamilton ico, N. Y., (Geoloey ct the. By
William J. Miller. Review by H.R. B. : :
Laws of Elastico-Viscous Flow, The. By A. A. Michelson
Lee, Willis T. Relations of Cretaceous Formations to the Rocky
Mountains in Colorado and New Mexico. Review by H. R. B.
Lee, W. T., R. W. Stone, H. S. Gale, and Others. Guidebook of the
Western United States. Part B, the Overland Route. Review
by. Wek. : LAA ee a ea Se
Lewis and Gilmer Counties IW. Val. ‘By David B. Reger. Review
by W.B.W. . MONET ea ena
Limestone Road Monerale of Tiseonai: By W. O. Hotchkiss and
Edward Steidtman. ReviewbyW.B.W. ..
Logan and Mingo Counties [W. Va.]. By R. V. Hennen ane DB. B.
Reger. Review by W. B. W. :
Loughlin, G. F. A Reconnaissance in the Ganvon Range West- Central
Utah. Review by V. O. T. : !
Lowe, E. N. Soils of Mississippi. Review by W: B. W.
Lull, Richard Swann. ‘Triassic Life of the Connecticut eee Re
view by C. H. E. si A Sl anentvae Wate aah apd 0) Sata
MacCurdy, George Grant. Archaeological Evidences of Man’s Antiq-
uity at Vero, Florida
McLeish, John. Mineral Broduetion: of Cannes for 1913. Review
by Des: f : : : :
MacMillan, W.D. Onthe Typothesis of Tastee :
Magmatic Sulfid Ores, The Origin of. By C. F. Tolman, ia
Austin F. Rogers. Review by E. A. Stephenson :
Malcolm, Wyatt. Oil and Gas Fields of Ontario and Quebec. Roe
Dysarts, atts a es ON ANE Ga ore ial| Te eee
PAGE
203
783
135
638
399
780
781
397
309
786
405
595
397
204
4O1
402
206
404
201
56
103
105
594
308
ee
— ee eS
INDEX TO VOLUME XXV
Manhattan Schist of Southeastern New York State and Its Associated
Igneous Rocks, The. By Charles Reinhard Fettke. Review by
~ Albert Johannsen
Man’s Antiquity, Archaeological Fvidences ae at Vero: Florida. By
George Grant MacCurdy é ;
Maritime Provinces, un Deposits of rhe By William F. Jennison:
Review by W.B.W. _.. :
Mathematical Theory of the Internal Friction and lnrarting Stren of
Rocks under Conditions of Stress Existing in the Interior of the
Earth, On the. By Louis Vessot King
Mather, Kirtley F. The Champlain Sea in the Lake Ontario Basin
Mauna Loa, Notes on the 1916 Eruption of. I and II. By Harry O.
Wood 2 : : ae : . : : :
Ibid. III and IV. : : 2 : : : :
Meinzer, O. E., and R. F. Hare. Geology and Water Resources of
Tularosa Basin, New Mexico. Review by C. W. T. : : ;
Michelson, A. A. The Laws of Elastico-Viscous Flow.
Middle Paleozoic Stratigraphy of the Central mes Mountain m Region,
The. I. By C. W. Tomlinson
Tide
III. Lea aug
Mikroscopische nterichase ales Wizlapersta teen iO By Georg
Berg. Review by A. D. B. : ie beanie
Miller, Gerret S., Jr. The Jaw of the Piltdown eae Review by
BL se
Miller, William J. Gooloey, si the rake Blencant Quadrangle Hamlon
Con N.Y., Review by H.R. B. .
Mineral Deposits of the Santa Rita and Patporia Moun hes
By Frank C. Schrader, with contributions by James M. Hill. Re-
view by C. W. T.
Mineral Production of Alabama for 1913, Statistics a ae By Charles
A. Abele. Review by A. D. B.
Mineral Production of Canada, for 1913. By Tobe Maen Review
by A.D: B. : Sates
Mineral Resources of Alaska! a A ide Brooks and Others: Review
by W. B. W. :
Mineral Resources of Alaska Ror 1Q14. Be AG TEL Brooks sad Others:
Review by H. R. B. :
Mineral Resources of New Menico: is Payette A. Tones! Review
by AG De B:
Mineralization in the Contact Nctimorahic Ore Deposits of aN Ofnicn
Mine, Prov. Nagato, Japan. By Takeo Kato. Review by
Albert Johannsen ‘ :
Minerals, An Arrangement of, according to Their @rctecnce: By
E. T. Wherry and S. T. Gordon. Review by A. D. B.
Mining Districts in Northeastern California and Northwestern N eta
Some. By James M. Hill. Review by C. W. T.
798 INDEX TO VOLUME XXV
Mississippian Seas in Illinois, Evidence Bearing on a Possible North-
eastward Extension of. By W. W. Davis . :
Missouri Bureau of Geology and Mines, Biennial Report of iBy H. A.
Buehler. Review by W. B.W. . :
Moffit, F. H. Geology of the Hanagita- Beenine: Region af alesis
Review by WB Wie
Moffit, Fred A. The Broad Pass Reson! Neale Revee by Et. R. B.
Montana Group of Northwestern Montana, The. By E. Stebinger.
Review by V. O. T. é : : ;
Moore, E.S. The Active Wolentoce i Nev ew Zealand i
Morgan, P.C.,and J. A. Bartrum. The Geology and Mineral Resources
of the Buller-Mokihinui Subdivision, Westport Division, New Zea-
land. Review by H. R. B. 2 LATINO ae
Mud-Cracks, Some Factors one the Development ¢ ae By E. M.
Kindle : : : ‘
Nephelite-Syenites of Haliburton County, Ontario. By Wilbur G. ee
Review by Albert Johannsen
Noble, L. F. The Shinumo Quadrangle. Review by W. B. W.
———. The Shinumo Quadrangle, Grand Canyon District, Arizona.
Review by H. R. B.
North Fork, Colorado, Geology and ‘coal Resourees 6b By Aw L:
Beekly. Review by COWS ees
Northwestern Oregon, Geology and Oil Drospects on By Cc: W. Wash:
burne. Review by W. B. W. ;
Norton, W.H. Studies for Students: A Glassacation of Bieccaet
Oil and Gas Fields of Ontario and Quebec. By Wyatt Malcolm.
Review by W.B.W.
Oinouye, Y. A Few Interesting Phenomend on the Frineon ci Ueu
Olentangy Shale of Central Ohio, Age and Stratigraphic Relations of the,
with Remarks on the Prout Limestone and So-called Olentangy
Shales of Northern Ohio. By Amadeus W. Grabau.
Osborn, Henry Fairfield. Review of the Pleistocene of PORT Ages
and Northern Africa. Review by H. R. B. : :
Palaechenéides Mioceanus, The Relationships of the Fossil Bird. By
Alexander Wetmore
Peat Resources of Wisconsin. ee F. W. lgele. Review by W. B. W :
Pebble Phosphates of Florida, The. By E. H. Sellards. Review by
W. B. W. : sgt Date hoe CE
Pennsylvanian Series in Miscou! Seale ai iin. By Henry Hinds
and F.C. Greene. Reviewby W.B.W. .
Perkins, G. H., and Others. Biennial Report of Vermont State Geol.
ogist. Review by W. B. W. :
Petrographic Evidence on the Origin of ate Gatahouls Sandstone 61
Texas. By MarcusI. Goldman. Review by Albert Johannsen
Petroleum and Natural Gas in Oklahoma. By C. W. Shannon and L. E.
Trout. Review by H. R. B. ; : : : : : : :
308
337
595
555
404
203
195
402
780
785
INDEX TO VOLUME XXV
799
PAGE
Petrological Abstracts and Reviews. By Albert Johannsen. . 4092, 587, 779
Petrological Notes on the Igneous Rocks of the Oki Islands. By S:
Kozu. Review by Albert Johannsen .
Philogeny and Classification of Reptiles, The. By S. W. Wallicton!
Physiography of the Beaverdell Map-Area and the Southern Part of the
Interior Plateaus of British Columbia. By Leopold Reinecke.
Review by H. R. B. ; : ‘ : ! . : :
Piltdown Man, The Jaw of the. By Gerret S. Miller, Jr. Review by
ERB. hii) CA aa Ve
Plagioclases, A Decimal Groupine of the: By F. C. Calkins
Pleistocene of Europe, Asia, and Northern Africa, Review of. By Henry
Fairfield Osborn. Review byHe Ro 'B: : Be a er hie
Pogue, Joseph E. The Turquoise. Review by A. D. ny
Quantitative Mineralogical Classification of Igneous Rocks, Suggestions
fora. By Albert Johannsen
Quaternary Deposits, The, at Vero, Florida, bad fe Wertebente Remains
Contained Therein. By Oliver P. Hay Pe ay Races
Recent Publications. . WMS NAGS OSO.
Red Iron Ores of East Miceneeee The. By Bree F. Burchard.
Review by V. O. T.
Reger, David B. Lewis and Gilmer counes [w. Val. Revie be
NV eB Wie (Our) 3 2
Reinecke, Leopold. He Cert OF fhe. Beaverdell Map Are and
the Southern Part of the Interior Plateaus of British Columbia.
Review by H. R. B.
Report of an Excursion to the ee Connell By J 5: Flett and
J.B. Hill. Review by Albert Johannsen
Reptiles, The Philogeny and Classification of. By S. W. “Williston
Residual Mississippian of Phelps County (Central Ozark Region),
Missouri, A Study of the Faunas of the. By Josiah Bridge ,
Reviews... he 983 1051307) 305 150400078
Richmond Group, at OWinchecter Ohio, Intraformational Pebbles in the.
By August F. Foerste. :
Rittman, W. F., C. B. Dutton, and E. W. iDyeee The vienulienie ak
Gasoline and Benzene-Toluene from Petroleum and other Hydro-
carbons. Review by A. D. B.
Robinson, Henry Hollister. The San Renee Voleame Field,
Arizona. Review by V. O. T.
Rocky Mountains in Colorado and New Mexico, Relanone of Cretarecus
Formations tothe. By Willis T. Lee. Review by H.R. B.
Rogers, Austin F. A Review of the Amorphous Minerals
‘Sanford, Samuel, and Ralph Stone. Useful Minerals of the United
States. Review by W. B. W. Dee ea ave ee ton hale AO ane
San Franciscan Volcanic Field, Arizona, The. By Henry Hollister
Robinson, WReEview Dy WelOs Weenie aoe G Rae sete Pieavaiitg vo ale
781
4IL
782
506
157
595
104
63
52
788
196
204
782
779
4II
558
782
289
207
199
595
515
IOI
199
800 INDEX TO VOLUME XXV
Santa Rita and Patagonia Mountains, Arizona, Mineral Deposits of the.
Byk ‘Frank C. Schrader, with contr ibutions a oe M. Hill. Re-
view by C. W. T. :
Satsop Formation of Oregon And Washington" The. By I eee iit:
Saunders, E. J. Coal Fields of Kittitas tae! [Wash.]. Review 2
W. B. W. ; :
Sayles, Robert W. The Sqraniin Tillite. Renee isl R. B.
Schrader, Frank C., with contributions by James M. Hill. Mineral
Deposits of the Santa Rita and Patagonia Mountains, Arizona.
Review by C. W. T.
Sellards, E. H. Note on the Deposits Goneanine Hunan Remar andl
Artifacts at Vero, Florida
On the Association of Human Rewuine ana Heenet Vere:
brates at Vero, Florida
The Pebble Phosphates of Blonde. Revie by W. B. W. :
Shannon, C. W.,and L. E. Trout. Petroleum and Natural Gas in Okla-
homa. Review by H.R. B. : :
Shinumo Quadrangle, Grand Caen District, Aoua The. ‘ByL. F.
Noble. Review by H. R. B.
Shinumo Quadrangle, The. By L. F. Noble. Review by We B. W.
Slate in the United States. By T. Nelson Dale and Others. Review
by W. B. W. ORE AR ORT A) NPL aa eS
Soils of Mississippi. By E.N. Lowe. Review by W.B.W. . :
Southern Colorado, Another Locality of Eocene Glaciation in. By
Wallace W. Atwood. PRC Cia) reataeee 2X EID. IN OI On ae
Southwestern Colorado, Contributions to the Stratigraphy of. By
Whitman Cross and Esper S. Larsen. Review by V. O. T. :
Southwestern Colorado, Eocene Glacial Deposits of. By Wallace W.
Atwood. Review by H.R. B. . : : ! d : : :
Squantum Tillite, The. By Robert W. Sayles. Review by H.R. B.
Stauffer, C. R. Devonian of Southwestern Ontario. Review by
Stebinger, E. The Montana Group of Northwestern Montana. Re-
views Dy Vitec kis eeu Mash AE 6s. Gh he tee saline ee er
Stephenson, E. A. Review of “The Origin of the Magmatic Sulfid
Ores,” by C. F. Tolman, Jr., and Austin F. Rogers
Stephenson, L. W. The Cretaceous-Eocene Contact in the Atlantic
and Gulf Coastal Plain. Review by V. O. T. 5;
Stratigraphy of Southwestern Colorado, Contributions to the. By
Whitman Cross and Esper S. Larsen. Reviewby V.O.T. ...
Stratigraphy of the Montana Group, with Special Reference to tHe
Position and Age of the Judith River Formation in North-Central
Montana, The. By C.F. Bowen. Review by V. O. T.
Stratigraphy of the Pennsylvanian Series in Missouri. ay Henry
Hinds and F. C. Greene. Review by W. B. W.
Studies for Students: A Classification of Breccias. By W. H. Nortant
Summary Report of Canadian Geological Survey. Review by W. B. W.
PAGE
INDEX TO VOLUME XXV
Supposed Oil-bearing Areas of South Australia. By Arthur Wade.
Review by W. B. W.
Symposium on the Age and Relations of ie Foreil Tetrvaaen Remains
Found at Vero, Florida .
Tolman, C. F., Jr., and Austin F. Rogers. The Origin of the Magmatic
Sulfid Ores. Review by E.A.Stephenson .
Tomlinson, C. W. The Middle Paleozoic ek aeee a a the Central
Rocky Mountain Region. I. :
100
Tdi
Transvaal, On the Geology of the Alkali Ror in aye Byes:
Brouwer
Triassic Life of the Counccree Valley By, Richard Seann Tull.
Review by C. H. E. :
Trowbridge, Arthur C. The History bi Devils Teles Wiiceonen
Tularosa Basin, New Mexico, Geology and Water Resources of. By
O. E. Meinzer and R. F. Hare. Review by: Cain ik: ee
Turquois, The. By Joseph E. Pogue. Review by A. D. B. a unses
Tyrrell, a B. Gold on the North Saskatchewan River. Review by
ia\a 1D Sie
The Gold- Basen Grave af Beaiee’ Co | Quebec. Review
bya. D. B.
Upper White River District, Yukon. By D. D. Cairnes. Review
by W. B. W.
Useful Minerals of the United States By Sammie Sanford ond Raion
Stone. Review by W. B. W.
Usu, A Few Interesting Phenomena on the eae ae Ey y. @ineue
Van Tuyl, Francis M. The Western Interior Geosyncline and Its
Bearing on the Origin and Distribution of the Coal Measures.
Vaughan, Thomas Wayland. On ee Pleistocene Human
Remains at Vero, Florida : ; ; ; : ‘
Vermont State Geologist, Biennial Report of By G. H. Perkins and
Others. Review by W. B. W. aj
Vero, Florida, Archaeological Evidences of Mente Ameautl ae By
George Grant MacCurdy Eee ra
Vero, Florida, Further Studies at. By Rolin T. @hamberia :
Vero, Florida, Interpretation of the Formations Containing Human
Bones at. By Roilin T. Chamberlin .
Vero, Florida, Note on the Deposits Containing Ea Remains and
Artifacts at. By E. H. Sellards . :
Vero, Florida, On Reported Pleistocene Human eewaine at. Ry
Thomas Wayland Vaughan
Vero, Florida, Preliminary Report on inde of Supposedly) Avacttan:
Human Remains at. By Ales Hrdlitka
Vero, Florida, Symposium on the Age and Relations ei the Rosa Teaver
Remains Found at . eS AN nem Canam a es mad SS ene
802 INDEX TO VOLUME XXV
PAGE
Vero, Florida, The Fossil Plants from. By Edward W. Berry. 661
Vero, Florida, The Quaternary Deposits at, and the Vertebrate Remains
Contained Therein. By Oliver P. Hay 52
Volcanoes of Japan, On the. By B. Koto. Review: by Albert t Johann:
sen 780
Volcanoes 5 NEG Zealand: The eae By E. S. Moore: 693
Wade, Arthur. Supposed Oil-bearing Areas of South Wastrala!
Review by W.B.W. . L205
Wallace, R.C. The Corrosive Action af Goran Bees in Y Menitobs 459
Washburne, C. W. Discussion of “Some Effects of Capillarity on Oil
Accumulation,” by A. W. McCoy. : 584
, Washburne, C.W. Geology and Oil Prospects of Nomhivestern Oreson!
Review by W.B.W. _. 102
Wegemann, Carroll H. The Coalville Coal Field: Utah. Review by
Caw. 203
Wentworth, Chester K. A Proposed Dip Broemictoe 489
Western Interior Geosyncline and Its Bearing on the Origin and Dishes
bution of the Coal Measures, The. By Francis M. Van Tuyl. 150
Wetmore, Alexander. The ere of the Fossil Bird Palae-
ochendides Mioceanus RE Ea em eden on SOS
Wherry, Eo: and: Ss.) 7: Gorden! An Arrangement of Minerals
according to Their Occurrence. Review by A. D. B. : : ae 2O7
Williams, M. Y. Arisaig-Antigonish District, Nova Scotia. Review
by W. B. W. 308
Williston, Samuel W. erbadosannne Coot a Howey Cotylosaur Reptile
from Texas 309
Williston S. W. The Bhiloseny: and Glaciation a Repeilest 411
. Willow Creek District, Alaska, The. By S. R. Capps. Review by
ERB sci" ; 785
Wood, Harry O. Nore) on the rot6 Eruption of Mauna Loa e228 AOm
Wyoming and McDowell Counties a ae By R. V. Hennen.
Review by W. B. W. ; : : ? : Hi BO
Yampa River, Colorado, The History of a Portion of, and Its Possible
Bearing on That of Green River. By E. T. Hancock. Review
by, VO: : anc ve MATa Ns iHieei eRe a XOGS
Yentna District, Alaska, The. By Stephen R. Capps. Review by
He RB. Shida GNtt hi aah iaai ae te aiid Wy a Ma ie a OS
Ziegler, Victor. Foothills Structure in Northern Colorado 75
A Manual of Style: New Fifth Edition
By THE STAFF OF THE UNIVERSITY OF CHICAGO PRESS
The increased value of the fifth edition will be evident to all who desire an Buehonibtaue guide on questions con-
cerning the style of a book or any other printed matter. An invaluable handbook for writers, editors, teachers,
librarians, ministers, lawyers, printers, proofreaders, etc.
270 pages, cloth; $1.50, postage extra (weight 1 Ib. 6 oz.)
Martin Luther and the Morning Hour in Europe
By Frank W. Gunsautus, Pastor of Central Church, Chicago, and President of the Armour
Institute of Technology
Commemorating the four-hundredth anniversary of the beginning of the German Reformation.
50 pages, paper; 25 cenis (postpaid 28 cents)
The Gospel of John
By Epcar J. GoopsPEED, Professor of Biblical and Patristic Greek, the University of Chicago
A careful historical study, throwing much light upon the authorship and purpose of the gospel of John, and the
part which it played in the thought and life of the early Christians.
44 pages, paper; 50 cents (postpaid 52 cents)
The Papers of the Bibliographical Society of America
Vol. XI, Nos. 3-4. July—October, 1917
Edited by Axsret G. S. JoSEPHSON of the John Crerar Memorial Library
74 pages, paper; $2:00 (postpaid $2.04)
Stellar Parallaxes Derived from Photographs Made with the 40-Inch
Refractor
(Publications of the Yerkes Observatory, Vol. IV, Part I)
By FREDERICK Stocum, S. ALFRED MITCHELL, OLIVER J. LEE, ALFRED H. Joy, and GEORGES
VAN BIESBROECK
: 74 pages and 2 plates, paper; $1.50, postage extra (weight 11 02.)
Contributions from Walker Museum
Vol. Il, No. 1. The Structure and Relationships of Diplocaulus
By Herman Dovurtuitt, Assistant Professor of Zodlogy, the University of Kansas
42 pages and 2 plates; 25 cents (postpaid 27 cents)
No. 2. Labidosaurus Cope, a Lower Permian Cotylosaur Reptile froth Texas
By SamMuet W. Wit.iston, Professor of Vertebrate Paleontology, the University of
Chicago
16 pages, paper; 25 cents (postpaid 28 cents)
No. 3. The Phylogeny and Classification of Reptiles
By SAmMurL W. WILLISTON, Professor of Vertebrate Paleontology, the University of
Chicago
14 pages, paper; 25 cents (postpaid 28 cents)
TO BE PUBLISHED IN DECEMBER
The Life of Lieutenant @eneral Chaffee
By Witi1am Harpine Carter, Major General, the United States Army. Commander of the
‘Central Department
“General Chaffee’ S career should be an inspiration to every young man whose tastes and qualifications suggest the
A Oe Bae 260 pages, 16 half-tone inserts, cloth; $2.50, postage extra ~
| Publications of the Members of the University of Chicago, 1902- 1916
Compiled on the Twenty-fifth Anniversary of the Foundation of the University, by a
COMMITTEE OF THE FACULTY :
526 pages, cloth; $1.50 (postpaid)
THE UNIVERSITY OF CHICAGO PRESS
CHICAGO ; - 5825 Ellis Avenue ILLINOIS
FINE INKS 4" ADHESIVES
For those who KNOW :
SELF STARTING
REMINGTON
TYPEWRITER
Saves from 15% to 25% of
every letter-writing hour
ee ee eee ee ee ee
Drawing Inks
Eternal Writing Ink
Engrossing Ink
2 2 9 Taurine Mucilage
rm | | pas gx ins Photo Mounter Paste
Drawing Board Paste
Liquid Paste
Office Paste
Vegetable Glue, Etc.
Are the Finest and Best Inks and Adhesives
Emancipate yourself from the use of corrosive and
ill-smelling inks and adhesives and adopt the
Higgins Inks and Adhesives. They willbea
revelation to you, they are so sweet, clean, well
put up, and withal so efficient.
At Dealers Generally
HIS new invention permits your
typist to keep her eyes onher copy.
The machinedoesn’t have to be looked
at, or the scale watched. The time
saving isautomatic, There is no other
typewriter like this. Fully protected
by Remington patents.
Descriptive folders mailed on request.
REMINGTON TYPEWRITER CO.
(Incorporated)
NEW YORK AND EVERYWHERE
CHAS. M. HIGGINS & CO., Mfrs.
Branches: Chicago, London
271 Ninth Street Brooklyn, N.Y.
Simple in Operation
Superior in Optics
Teachers, lecturers, instructors in classroom
or laboratory, find that all the requirements
_ of a projection lantern are completely met by the
Bausch |omb
BALOPTICON
THE PERFECT STEREOPTICON
Its superiority of optical equipment, giving remarkable clearness to the corners of the image, its new and
powerful Mazda lamp equipment, and its excellent mechanical construction combine with its ease of operation
to make it an ideal means of graphic presentation for frequent use.
Model C—for slides. Has new 400-watt Mazda lamp, giving illumination superior to that of any similar
instrument on the market. Price, $35.00 up.
Combined Model—for both slides and opaque objects. Equipped with 1000-watt lamp giving superior
illumination to the 35-ampere, A.C. arc lamp—vet using only 9 amperes. Price, $120.
Other mode's from $26.50 up.
Write for interesting descriptive matter and illustrated price list.
Bausch ¢%5 lomb Optical G.
654 ST, PAUL STREET, ROCHESTER, N.Y.
NEW YORK, WASHINGTON, CHICAGO, SAN FRANCISCO
Leading American Makers of Microscopes, Projection Apparatus, (Balopticons) Photographic and Ovhthaimic Baas :
Stereo-Prism Binoculars, and other High-Grade Optical Products.
shi
LOHR a RR Lt
DG Ga ta
. i
Foca
aK Si)
vi) i
Hy
el hon
,
ty
ve
i A
oe,
A
Peas :
Botlt IB
ess s ARg, G..
rates
fd
\ ae
Xe Aim by
iam ener
%. =
Ue
eel
ch
Peni
eT
cel
a
Ss,
thet
ee
5
ie
De
=
OED ME
se Ed,
hae" |
es
sae),
at
“Ry,
hy
3
ee
ie
+
et
; .G “h ei
it Papo e
Capa gOS |
*s va by
ee
apenas
==
getty
ZB ue Caner rs ss
te SRE Eze |
2 ieee teal is
2yn0 Te
NO
isaeeiagy
LED Bag
LAL
rama ee ql
2.
ier
is
a
TN.
ens
ra
gt oar
eee
eo
eel ;
DASAr po
arenas
aaa
fie
Swe
fis
es
a
Phe He
nN
Ra BF Nee ~
Sef lg hey
o ot Lay ch
fe
Wears
Pe eis
, i, fea ats,
(a
©)
=
s
=
ih
3 9088 01367 0146
|
|
|
ti
|
Mh
|
SMITHSONIAN INSTITUTION LIBRARIE
iM
ee ee
SPS
Si Ra
ee ete ae
YR ERA PUT Es stot
IS
Live Music
Archive
Librivox
Free Audio
Metropolitan Museum
Cleveland
Museum of Art
Internet
Arcade
Console Living Room
Open Library
American
Libraries
TV News
Understanding
9/11